{ "generated_at": "2026-06-25", "counts": { "core": 150, "candidate": 332, "all": 803 }, "core": [ { "id": "phase2-pal-aip-observability-overview-2026", "title": "AIP observability: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-observability/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing AIP observability features for AIP and Ontology workflow executions, including metrics, tracing, logs, execution history, and Workflow Lineage integration.", "key_claims": [ "AIP observability provides visibility into AIP and Ontology workflow executions through metrics, tracing, logs, and execution history.", "Workflow Lineage integrates observability across applications, workflows, and products built with AIP and the Ontology.", "Runtime observability is part of governing AI workflows, not just offline model evaluation.", "AIP observability provides visibility into AIP and Ontology workflow executions.", "It includes metrics, tracing, logs, and execution history.", "It is integrated into Workflow Lineage for cross-functional monitoring." ], "ontology_relevance": "Shows how operational ontology workflows can be monitored and traced.", "ai_relevance": "Important source for production AI observability and runtime governance.", "palantir_relevance": "Primary source for Palantir's claim that AIP/Ontology workflows are observable in production.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-observability", "core", "governance", "logs", "metrics", "palantir", "tracing", "workflow-lineage" ], "triage_tier": "core", "triage_score": 142 }, { "id": "phase2-pal-aip-evals-ontology-edits-2026", "title": "AIP Evals: Evaluate Ontology edits", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-evals/ontology-edits", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation stating that when evaluated Logic functions create, edit, or delete ontology objects, each test case executes in an Ontology simulation so the actual Ontology remains unchanged.", "key_claims": [ "Write-capable AI functions can be evaluated in ontology simulations before affecting real ontology state.", "AIP Evals treats ontology edits as a special case requiring isolation during testing.", "Simulation is a concrete control for non-deterministic LLM workflows with operational side effects.", "AIP Logic functions that result in ontology edits require custom evaluation functions or intermediate parameters.", "Evaluation functions must return Boolean or numeric values, or metrics in a struct.", "Ontology-editing workflows need specialized evaluation because ordinary outputs may not capture state changes." ], "ontology_relevance": "Shows how ontology edits and operational state changes can be sandboxed for evaluation.", "ai_relevance": "High-value evidence for evaluating action-capable AI workflows before deployment.", "palantir_relevance": "Core Palantir source for safe testing of ontology writebacks by AIP Logic.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-evals", "core", "custom-evaluation", "governed-action", "ontology-edits", "palantir", "simulation", "validation", "write-capable-agents" ], "triage_tier": "core", "triage_score": 130 }, { "id": "core-palantir-ontology-overview-2026", "title": "Ontology building: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation defining Palantir Ontology as an operational layer over datasets, virtual tables, and models, with semantic and kinetic elements.", "key_claims": [ "Palantir Ontology sits above integrated digital assets and connects them to real-world counterparts.", "The Ontology includes semantic elements and kinetic elements such as actions, functions, and dynamic security.", "The Ontology sits on top of integrated digital assets and connects them to real-world counterparts.", "The Ontology can serve as a digital twin containing semantic elements and kinetic elements.", "Action types and functions define the kinetic parts of organizational change.", "The Ontology sits above integrated digital assets and connects them to real-world counterparts.", "The Ontology is positioned as a digital twin of organizational operations.", "Semantic elements describe entities and relationships, while kinetic elements such as action types and functions define ways to change operations.", "The Ontology sits on top of integrated digital assets and maps them to real-world counterparts.", "Palantir distinguishes semantic elements from kinetic elements.", "Action types and functions define governed ways to change operations." ], "ontology_relevance": "Primary source for Palantir's product-specific ontology definition.", "ai_relevance": "Defines the operational substrate that AIP uses to connect AI to enterprise workflows.", "palantir_relevance": "Core Palantir source.", "quality_signal": "official_docs", "retrieval_tags": [ "actions", "core", "digital-twin", "foundry", "kinetic-elements", "ontology", "operational-layer", "palantir", "semantic-elements" ], "triage_tier": "core", "triage_score": 130 }, { "id": "oa-https-doi-org-10-1145-3447772", "title": "Knowledge Graphs", "authors_or_org": "Aidan Hogan, Eva Blomqvist, Michael Cochez, Claudia d’Amato, Gerard de Melo, Claudio Gutiérrez, Sabrina Kirrane, José Emilio Labra Gayo", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1145/3447772", "doi_or_identifier": "10.1145/3447772", "venue_or_site": "ACM Computing Surveys", "abstract_or_summary": "In this article, we provide a comprehensive introduction to knowledge graphs, which have recently garnered significant attention from both industry and academia in scenarios that require exploiting diverse, dynamic, large-scale collections of data. After some opening remarks, we motivate and contrast various graph-based data models, as well as languages used to query and validate knowledge graphs. We explain how knowledge can be represented and extracted using a combination of deductive and inductive techniques. We conclude with high-level future research directions for knowledge graphs.", "key_claims": [ "Knowledge graphs combine graph-structured data with semantics, identifiers, and sometimes formal constraints.", "KG ecosystems include representation, querying, reasoning, construction, quality, and embeddings.", "Knowledge graphs combine graph-structured data with schema, identity, semantics, and sometimes formal reasoning.", "Graph embeddings and symbolic reasoning address different aspects of graph-based knowledge.", "Knowledge graph construction and refinement involve extraction, integration, quality control, and curation.", "Knowledge graphs combine graph-structured data with semantics and context.", "KGs support integration, search, analytics, and reasoning over heterogeneous sources.", "Construction, quality, and maintenance are central technical challenges." ], "ontology_relevance": "Connects ontologies to the wider knowledge graph ecosystem.", "ai_relevance": "Core survey for KG-enabled AI, retrieval, reasoning, and embeddings.", "palantir_relevance": "Useful for comparing Palantir-style ontology to broader knowledge graph theory and practice.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "ai", "commercial", "construction", "core", "data-integration", "embeddings", "knowledge-graph", "knowledge-graphs", "openalex", "rdf", "reasoning", "semantic-web", "survey" ], "triage_tier": "core", "triage_score": 127 }, { "id": "phase2-nist-genai-profile-2024", "title": "Artificial Intelligence Risk Management Framework: Generative Artificial Intelligence Profile", "authors_or_org": "National Institute of Standards and Technology", "year": 2024, "source_type": "technical_report", "bucket": "technical", "url": "https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.600-1.pdf", "doi_or_identifier": "nist ai 600-1", "venue_or_site": "NIST", "abstract_or_summary": "NIST profile applying AI RMF concepts to generative AI, including risks around confabulation, data privacy, harmful content, and information integrity.", "key_claims": [ "Generative AI systems create specific risks including confabulation, data privacy leakage, harmful content, and information integrity failures.", "Risk controls should be mapped to specific use contexts rather than treated as generic model properties.", "Ontology-grounded RAG and agent systems need evaluation for grounding, provenance, misuse, privacy, and action consequences." ], "ontology_relevance": "Useful for framing ontology as part of the control layer for generative AI risk management.", "ai_relevance": "Authoritative generative AI governance source for RAG and agent risk discussion.", "palantir_relevance": "Provides non-vendor risk criteria for assessing AIP and ontology-mediated agent workflows.", "quality_signal": "official_standard", "retrieval_tags": [ "core", "evaluation", "generative-ai", "governance", "hallucination", "nist", "privacy", "risk" ], "triage_tier": "core", "triage_score": 126 }, { "id": "core-w3c-rdf-11-concepts", "title": "RDF 1.1 Concepts and Abstract Syntax", "authors_or_org": "W3C", "year": 2014, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/rdf11-concepts", "doi_or_identifier": "w3c recommendation 25 february 2014", "venue_or_site": "W3C", "abstract_or_summary": "W3C standard defining RDF's graph data model, IRIs, literals, triples, and datasets.", "key_claims": [ "RDF represents information as graph triples.", "IRIs and literals provide a Web-scale naming and value model for semantic data.", "RDF represents information as graph-based triples.", "IRIs provide global identifiers for resources and properties.", "RDF datasets support named graphs and graph-based integration.", "RDF represents information as subject-predicate-object triples.", "IRIs provide global identifiers for resources and predicates.", "RDF graphs support interoperable data integration." ], "ontology_relevance": "Base data model for Semantic Web ontologies and linked data.", "ai_relevance": "Provides graph representation substrate for symbolic and hybrid AI systems.", "palantir_relevance": "Relevant as a standards-based contrast to proprietary enterprise ontology graph models.", "quality_signal": "official_standard", "retrieval_tags": [ "core", "graph-data", "linked-data", "rdf", "semantic-web", "standard", "w3c" ], "triage_tier": "core", "triage_score": 126 }, { "id": "w3c_shacl_2017", "title": "Shapes Constraint Language (SHACL)", "authors_or_org": "W3C RDF Data Shapes Working Group", "year": 2017, "source_type": "docs", "bucket": "technical", "url": "https://www.w3.org/TR/shacl", "doi_or_identifier": "w3c:rec-shacl-20170720", "venue_or_site": "World Wide Web Consortium Recommendation", "abstract_or_summary": "Defines SHACL for validating RDF graphs against shapes graphs that express constraints over nodes and properties.", "key_claims": [ "SHACL separates data graphs from shapes graphs used for validation.", "Constraint validation can check cardinality, datatypes, classes, paths, and custom conditions.", "RDF graph quality can be tested before downstream use.", "SHACL validates RDF graph data against declared constraints.", "Shape validation supports data quality and application-level requirements.", "Constraint checking complements OWL reasoning.", "SHACL gives RDF and ontology systems a standard validation layer beyond open-world inference.", "Shape graphs can encode structural and value constraints that make semantic data operationally checkable.", "Validation is a missing bridge between formal ontology and production data quality controls." ], "ontology_relevance": "Key standard for validating ontology-backed graph data and enforcing data-shape expectations.", "ai_relevance": "Useful guardrail for AI-generated graph updates, extraction pipelines, and agent tool outputs.", "palantir_relevance": "Relevant to operational data quality checks around enterprise ontology objects and relationships.", "quality_signal": "official_standard", "retrieval_tags": [ "constraints", "core", "data-quality", "ontology-governance", "rdf", "semantic-governance", "semantic-web", "shacl", "validation", "w3c" ], "triage_tier": "core", "triage_score": 126 }, { "id": "core-palantir-aip-overview-2026", "title": "AIP overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official AIP documentation describing AI builder tools for workflows, agents, functions, evals, and applications built on top of the Ontology.", "key_claims": [ "AIP builder tools support AI-powered workflows, agents, and functions on top of the Ontology.", "AIP integrates LLMs with platform security, observability, and developer tooling.", "AIP builder tools include AIP Logic, AIP Chatbot Studio, and AIP Evals.", "AIP enables AI workflows, agents, and functions on top of the Ontology and developer toolchain.", "AIP integrates generative AI into Palantir's application environment with security, audit, and resource management.", "AIP integrates generative AI into Palantir's application environment with governance, audit, and resource management concepts.", "AIP connects AI with data and operations.", "AIP is designed to drive automation across operational processes.", "The platform targets both developers and frontline users." ], "ontology_relevance": "Shows Palantir's ontology as the context/tool layer for AI systems.", "ai_relevance": "Primary source for Palantir's AI platform architecture claims.", "palantir_relevance": "Core Palantir AIP source.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "ai-platform", "aip", "aip-evals", "aip-logic", "automation", "core", "generative-ai", "llm", "ontology", "operations", "palantir" ], "triage_tier": "core", "triage_score": 124 }, { "id": "oa-https-doi-org-10-1109-tnnls-2021-3070843", "title": "A Survey on Knowledge Graphs: Representation, Acquisition, and Applications", "authors_or_org": "Shaoxiong Ji, Shirui Pan, Erik Cambria, Pekka Marttinen, Philip S. Yu", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://ieeexplore.ieee.org/document/9416312", "doi_or_identifier": "10.1109/tnnls.2021.3070843", "venue_or_site": "IEEE Transactions on Neural Networks and Learning Systems", "abstract_or_summary": "Human knowledge provides a formal understanding of the world. Knowledge graphs that represent structural relations between entities have become an increasingly popular research direction toward cognition and human-level intelligence. In this survey, we provide a comprehensive review of the knowledge graph covering overall research topics about: 1) knowledge graph representation learning; 2) knowledge acquisition and completion; 3) temporal knowledge graph; and 4) knowledge-aware applications and summarize recent breakthroughs and perspective directions to facilitate future research. We propose a full-view categorization and new taxonomies on these topics. Knowledge graph embedding is organized from four aspects of representation space, scoring function, encoding models, and auxiliary information. For knowledge acquisition, especially knowledge graph completion, embedding methods, path inference, and logical rule reasoning are reviewed. We further explore several emerging topics, including metarelational learning, commonsense reasoning, and temporal knowledge graphs. To facilitate future research on knowledge graphs, we also provide a curated collection of data sets and open-source libraries on different tasks. In the end, we have a thorough outlook on several promising research directions.", "key_claims": [ "Knowledge graph research spans representation learning, knowledge acquisition, completion, fusion, and reasoning.", "KGs support search, recommendation, question answering, and explainability.", "KG research spans representation, acquisition, fusion, completion, and application.", "Embedding and neural methods help with incomplete KGs but do not replace symbolic semantics.", "Applications include question answering, recommendation, and information retrieval." ], "ontology_relevance": "Shows where formal ontologies meet modern KG machine learning.", "ai_relevance": "High-value map of KG methods for AI applications.", "palantir_relevance": "Useful to compare enterprise ontology with broader KG AI infrastructure.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "applications", "core", "foundational", "kg-completion", "knowledge-acquisition", "knowledge-graph", "ontology_ai", "openalex", "representation-learning", "survey" ], "triage_tier": "core", "triage_score": 121 }, { "id": "phase2-nist-ai-rmf-1-0-2023", "title": "Artificial Intelligence Risk Management Framework (AI RMF 1.0)", "authors_or_org": "National Institute of Standards and Technology", "year": 2023, "source_type": "technical_report", "bucket": "technical", "url": "https://www.nist.gov/itl/ai-risk-management-framework", "doi_or_identifier": "nist ai rmf 1.0", "venue_or_site": "NIST", "abstract_or_summary": "Official NIST framework for managing AI risks through govern, map, measure, and manage functions across the AI lifecycle.", "key_claims": [ "AI risk management requires governance, mapping, measurement, and management across the system lifecycle.", "Trustworthy AI characteristics include validity, reliability, safety, security, resilience, accountability, transparency, explainability, privacy, and fairness.", "Ontology-backed AI systems need explicit risk ownership, measurement, monitoring, and documentation, not only better retrieval." ], "ontology_relevance": "Provides governance vocabulary for ontology-backed AI systems that become operational infrastructure.", "ai_relevance": "Authoritative AI risk framework relevant to LLM, RAG, and agent deployment governance.", "palantir_relevance": "Useful external benchmark for evaluating Palantir AIP governance claims around security, observability, and controls.", "quality_signal": "official_standard", "retrieval_tags": [ "ai-risk-management", "core", "evaluation", "governance", "nist", "policy", "trustworthy-ai" ], "triage_tier": "core", "triage_score": 120 }, { "id": "phase2-w3c-prov-o-2013", "title": "PROV-O: The PROV Ontology", "authors_or_org": "W3C Provenance Working Group", "year": 2013, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/prov-o", "doi_or_identifier": null, "venue_or_site": "W3C", "abstract_or_summary": "W3C Recommendation expressing the PROV data model as an OWL2 ontology for representing provenance across systems and contexts.", "key_claims": [ "PROV-O provides a standard ontology for entities, activities, agents, and their provenance relations.", "Provenance is essential when knowledge graphs are generated, transformed, or used as evidence.", "AI knowledge bases need provenance if generated assertions must remain auditable." ], "ontology_relevance": "Authoritative provenance ontology for graph lineage, evidence, and transformation history.", "ai_relevance": "Directly supports trustworthy RAG, graph construction, agent traceability, and auditability.", "palantir_relevance": "Useful external standard for comparing Palantir lineage, observability, and audit claims.", "quality_signal": "official_standard", "retrieval_tags": [ "audit", "core", "lineage", "owl", "prov-o", "provenance", "trustworthy-ai", "w3c" ], "triage_tier": "core", "triage_score": 120 }, { "id": "phase2-eu-ai-act-2024", "title": "Regulation (EU) 2024/1689 laying down harmonised rules on artificial intelligence (Artificial Intelligence Act)", "authors_or_org": "European Parliament and Council of the European Union", "year": 2024, "source_type": "standard", "bucket": "technical", "url": "https://eur-lex.europa.eu/eli/reg/2024/1689/oj", "doi_or_identifier": "regulation (eu) 2024/1689", "venue_or_site": "Official Journal of the European Union", "abstract_or_summary": "EU regulation establishing rules for AI systems, including risk classification, documentation, transparency, human oversight, logging, post-market monitoring, and governance requirements.", "key_claims": [ "High-risk AI systems require risk management, data governance, technical documentation, logging, transparency, human oversight, accuracy, robustness, and cybersecurity controls.", "AI governance is increasingly a legal and operational requirement, not only an ethical preference.", "Operational ontology systems that drive consequential decisions may need auditability, role controls, documentation, and monitoring aligned with legal risk categories." ], "ontology_relevance": "Provides legal governance context for ontology-backed AI decision and action systems.", "ai_relevance": "Important regulatory source for AI systems deployed in public or high-risk contexts.", "palantir_relevance": "Useful for discussing Palantir public-sector deployments and AI governance expectations in regulated environments.", "quality_signal": "official_standard", "retrieval_tags": [ "core", "eu-ai-act", "governance", "human-oversight", "logging", "public-sector", "regulation", "risk-management" ], "triage_tier": "core", "triage_score": 120 }, { "id": "core-gao-2023-rag-survey", "title": "Retrieval-Augmented Generation for Large Language Models: A Survey", "authors_or_org": "Yunfan Gao; Yun Xiong; Xinyu Gao; Kangxiang Jia; Jinliu Pan; Yuxi Bi; Yi Dai; Jiawei Sun; Meng Wang; Haofen Wang", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2312.10997", "doi_or_identifier": "arxiv:2312.10997", "venue_or_site": "arXiv", "abstract_or_summary": "Survey of RAG paradigms, retrieval, augmentation, generation, evaluation, and applications for LLMs.", "key_claims": [ "RAG systems depend on retrieval quality, knowledge organization, and generation control.", "Advanced RAG includes query transformation, reranking, iterative retrieval, and evaluation.", "RAG reduces reliance on static parametric memory and supports external evidence.", "RAG quality depends on data indexing, retrieval, reranking, and generation alignment.", "Evaluation must include retrieval quality, answer correctness, faithfulness, and robustness.", "RAG quality depends on indexing, retrieval, reranking, generation, and evaluation as a full pipeline.", "Advanced RAG includes query rewriting, routing, iterative retrieval, and feedback loops.", "Evaluation must cover retrieval quality, answer faithfulness, robustness, and task utility." ], "ontology_relevance": "Shows why ontology-guided indexing and structured retrieval matter.", "ai_relevance": "Recent map of RAG methods for LLM applications.", "palantir_relevance": "Relevant to designing ontology-backed AIP-style retrieval and evidence pipelines.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "advanced-rag", "core", "evaluation", "llm", "rag", "retrieval", "retrieval-pipeline", "survey" ], "triage_tier": "core", "triage_score": 120 }, { "id": "arxiv-2306-08302v3", "title": "Unifying Large Language Models and Knowledge Graphs: A Roadmap", "authors_or_org": "Shirui Pan, Linhao Luo, Yufei Wang, Chen Chen, Jiapu Wang, Xindong Wu", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2306.08302", "doi_or_identifier": "10.1109/tkde.2024.3352100", "venue_or_site": "arXiv", "abstract_or_summary": "Large language models (LLMs), such as ChatGPT and GPT4, are making new waves in the field of natural language processing and artificial intelligence, due to their emergent ability and generalizability. However, LLMs are black-box models, which often fall short of capturing and accessing factual knowledge. In contrast, Knowledge Graphs (KGs), Wikipedia and Huapu for example, are structured knowledge models that explicitly store rich factual knowledge. KGs can enhance LLMs by providing external knowledge for inference and interpretability. Meanwhile, KGs are difficult to construct and evolving by nature, which challenges the existing methods in KGs to generate new facts and represent unseen knowledge. Therefore, it is complementary to unify LLMs and KGs together and simultaneously leverage their advantages. In this article, we present a forward-looking roadmap for the unification of LLMs and KGs. Our roadmap consists of three general frameworks, namely, 1) KG-enhanced LLMs, which incorporate KGs during the pre-training and inference phases of LLMs, or for the purpose of enhancing understanding of the knowledge learned by LLMs; 2) LLM-augmented KGs, that leverage LLMs for different KG tasks such as embedding, completion, construction, graph-to-text generation, and question answering; and 3) Synergized LLMs + KGs, in which LLMs and KGs play equal roles and work in a mutually beneficial way to enhance both LLMs and KGs for bidirectional reasoning driven by both data and knowledge. We review and summarize existing efforts within these three frameworks in our roadmap and pinpoint their future research directions.", "key_claims": [ "KGs can improve LLM factuality, reasoning, interpretability, and domain adaptation.", "LLMs can help construct, complete, query, and reason over KGs.", "LLMs can support KG construction, completion, alignment, and question answering.", "Future systems require tight integration of parametric and symbolic knowledge.", "LLMs can help construct and query KGs, while KGs can ground and constrain LLMs.", "The hybrid frontier is bidirectional: KG-enhanced LLMs and LLM-augmented KGs.", "Evaluation must consider factuality, reasoning, interpretability, and dynamic updates.", "LLMs can assist KG construction, completion, alignment, and question answering.", "Future systems will likely combine parametric and symbolic knowledge bidirectionally." ], "ontology_relevance": "Central source for mutual reinforcement between ontologies/KGs and LLMs.", "ai_relevance": "High-value recent synthesis for LLM+KG research agenda.", "palantir_relevance": "Useful neutral frame for Palantir-style ontology plus LLM operational AI.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "arxiv", "core", "grounding", "hybrid-ai", "knowledge-graph", "large-language-models", "llm", "llm-kg", "ontology_ai", "reasoning", "roadmap", "survey", "tkde" ], "triage_tier": "core", "triage_score": 120 }, { "id": "phase2-pal-ontology-mcp-overview-2026", "title": "Ontology MCP (OMCP) overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-mcp/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing Ontology MCP as a Developer Console feature that exposes application ontology resources as MCP tools for external AI agents and systems.", "key_claims": [ "Ontology MCP exposes object types, action types, and query functions as MCP tools.", "External agents can discover and use ontology resources to read objects, execute actions, and query data.", "Palantir positions OMCP as a safe read/write bridge between external agents and the Ontology." ], "ontology_relevance": "Primary source for Palantir's ontology-as-agent-tool-interface design.", "ai_relevance": "Direct evidence of ontology resources being exposed to external AI agents via MCP.", "palantir_relevance": "Core Palantir source for OMCP and agent integration.", "quality_signal": "official_docs", "retrieval_tags": [ "action-type", "agents", "core", "mcp", "object-type", "ontology-mcp", "palantir", "tool-use" ], "triage_tier": "core", "triage_score": 118 }, { "id": "core-edge-2024-graphrag", "title": "From Local to Global: A Graph RAG Approach to Query-Focused Summarization", "authors_or_org": "Darren Edge; Ha Trinh; Newman Cheng; Joshua Bradley; Alex Chao; Apurva Mody; Steven Truitt; Jonathan Larson", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2404.16130", "doi_or_identifier": "arxiv:2404.16130", "venue_or_site": "arXiv / Microsoft Research", "abstract_or_summary": "Presents GraphRAG, using graph-based communities and summaries to support global question answering over private datasets.", "key_claims": [ "Graph-structured indexes can improve synthesis over large corpora beyond local chunk retrieval.", "Community summaries provide higher-level retrieval units for broad questions.", "Graph structure can improve query-focused summarization over large corpora.", "Community detection and summaries support global questions not handled well by local chunk retrieval.", "Graph construction quality is a key dependency.", "Graph-based indexes can improve answers to global sensemaking questions that naive chunk retrieval handles poorly.", "Community summaries create retrieval units above local chunks, supporting broad query-focused summarization.", "Generated graph indexes require attention to entity extraction quality, provenance, and evaluation.", "Graph communities create retrieval units above raw document chunks.", "Global questions over large corpora need graph-level summaries, not only nearest-neighbor passages.", "Generated graphs are useful but make extraction quality and provenance central system risks.", "Plain chunk retrieval struggles with global questions over a whole corpus.", "Entity graphs and community summaries provide retrieval units above raw passages.", "Graph construction and summarization quality become central dependencies for answer quality." ], "ontology_relevance": "Supports graph/index architecture for the local ontology research knowledge base.", "ai_relevance": "Relevant to graph-augmented RAG and multi-hop synthesis.", "palantir_relevance": "Very relevant to enterprise document-to-object graph retrieval and operational summarization.", "quality_signal": "primary_source", "retrieval_tags": [ "community-summary", "core", "enterprise-search", "evaluation", "graphrag", "knowledge-graph", "microsoft", "private-data", "provenance", "query-focused-summarization", "rag", "retrieval", "summarization" ], "triage_tier": "core", "triage_score": 114 }, { "id": "phase2-iso-iec-42001-2023", "title": "ISO/IEC 42001:2023 Artificial intelligence - Management system", "authors_or_org": "ISO/IEC", "year": 2023, "source_type": "standard", "bucket": "technical", "url": "https://www.iso.org/standard/81230.html", "doi_or_identifier": "iso/iec 42001:2023", "venue_or_site": "ISO/IEC", "abstract_or_summary": "International standard specifying requirements for establishing, implementing, maintaining, and continually improving an AI management system within organizations.", "key_claims": [ "AI management systems require organizational processes for responsibility, risk, lifecycle controls, monitoring, and continual improvement.", "Governance must be embedded in management systems, not left to individual model or prompt choices.", "Ontology and knowledge graph programs should be governed as AI infrastructure when they support automated decisions or actions." ], "ontology_relevance": "Frames ontology/KG infrastructure as part of organizational AI management and lifecycle governance.", "ai_relevance": "Authoritative standard for AI management systems and organizational controls.", "palantir_relevance": "Useful benchmark for enterprise AIP/Foundry governance, ownership, and lifecycle controls.", "quality_signal": "official_standard", "retrieval_tags": [ "ai-management-system", "core", "governance", "iso", "lifecycle", "policy", "risk-management" ], "triage_tier": "core", "triage_score": 114 }, { "id": "w3c_skos_reference_2009", "title": "SKOS Simple Knowledge Organization System Reference", "authors_or_org": "W3C Semantic Web Deployment Working Group", "year": 2009, "source_type": "docs", "bucket": "technical", "url": "https://www.w3.org/TR/skos-reference", "doi_or_identifier": "w3c rec-skos-reference-20090818", "venue_or_site": "World Wide Web Consortium Recommendation", "abstract_or_summary": "Defines SKOS as a model for thesauri, classification schemes, taxonomies, subject headings, and other knowledge organization systems.", "key_claims": [ "SKOS models concepts, labels, notation, semantic relations, and concept schemes.", "SKOS supports lightweight knowledge organization without full OWL commitment.", "SKOS can publish controlled vocabularies as linked data.", "SKOS makes controlled vocabularies and taxonomies publishable and linkable as Web data.", "SKOS is lighter-weight than OWL but valuable for indexing, concept schemes, labels, and semantic retrieval.", "Many AI retrieval systems need SKOS-like concept organization before they need full formal ontology." ], "ontology_relevance": "Important lightweight standard adjacent to formal ontology, useful for taxonomies and controlled vocabularies.", "ai_relevance": "Useful for retrieval labels, semantic search, and controlled vocabulary grounding in AI applications.", "palantir_relevance": "Relevant where enterprise teams need pragmatic taxonomy layers before or alongside stronger ontology commitments.", "quality_signal": "official_standard", "retrieval_tags": [ "controlled-vocabulary", "core", "retrieval", "semantic-web", "skos", "standard", "taxonomy", "w3c" ], "triage_tier": "core", "triage_score": 114 }, { "id": "core-w3c-sparql-11-query", "title": "SPARQL 1.1 Query Language", "authors_or_org": "W3C SPARQL Working Group", "year": 2013, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/sparql11-query", "doi_or_identifier": "w3c:rec-sparql11-query-20130321", "venue_or_site": "W3C", "abstract_or_summary": "W3C standard query language for RDF graphs, including graph pattern matching and result forms.", "key_claims": [ "SPARQL provides standardized graph query over RDF datasets.", "Graph pattern matching supports retrieval across linked semantic data.", "SPARQL queries RDF graphs through graph patterns.", "Property paths and federation allow traversal and distributed querying.", "Standard graph querying is necessary for interoperable RDF systems.", "SPARQL enables pattern matching over RDF graph data.", "Graph queries can express joins, filters, aggregates, and optional patterns.", "Standard querying is essential for interoperable semantic systems." ], "ontology_relevance": "Query layer for RDF/OWL knowledge bases.", "ai_relevance": "Important for precise retrieval and symbolic querying in AI knowledge systems.", "palantir_relevance": "Relevant as a standards-based reference point for querying enterprise semantic layers.", "quality_signal": "official_standard", "retrieval_tags": [ "core", "graph-query", "knowledge-graph", "query", "query-language", "rdf", "semantic-web", "sparql", "standard", "standards", "w3c" ], "triage_tier": "core", "triage_score": 114 }, { "id": "oa-https-doi-org-10-1006-ijhc-1995-1081", "title": "Toward principles for the design of ontologies used for knowledge sharing?", "authors_or_org": "Thomas Gruber", "year": 1995, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1006/ijhc.1995.1081", "doi_or_identifier": "10.1006/ijhc.1995.1081", "venue_or_site": "International Journal of Human-Computer Studies", "abstract_or_summary": "Develops design principles for shareable ontologies, including clarity, coherence, extendibility, minimal encoding bias, and minimal ontological commitment.", "key_claims": [ "Ontology design should balance clarity, coherence, extendibility, encoding neutrality, and minimal commitment.", "Explicit design criteria make ontologies easier to share and reuse.", "Ontology design should make intended meanings clear and coherent.", "A reusable ontology should avoid unnecessary representation-specific commitments.", "Minimal ontological commitment can improve reuse across applications." ], "ontology_relevance": "Core source for ontology engineering evaluation criteria.", "ai_relevance": "Provides principles for building reusable semantic layers for AI systems.", "palantir_relevance": "Relevant to balancing reusable enterprise object semantics against application-specific operational constraints.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "clarity", "coherence", "core", "foundational", "gruber", "knowledge-sharing", "minimal-commitment", "ontology-design", "openalex" ], "triage_tier": "core", "triage_score": 114 }, { "id": "oa-https-doi-org-10-3233-sw-140134", "title": "DBpedia – A large-scale, multilingual knowledge base extracted from Wikipedia", "authors_or_org": "Jens Lehmann, Robert Isele, Max Jakob, Anja Jentzsch, Dimitris Kontokostas, Pablo N. Mendes, Sebastian Hellmann, Mohamed Morsey", "year": 2015, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/SW-140134", "doi_or_identifier": "10.3233/sw-140134", "venue_or_site": "Semantic Web", "abstract_or_summary": "The DBpedia community project extracts structured, multilingual knowledge from Wikipedia and makes it freely available on the Web using Semantic Web and Linked Data technologies. The project extracts knowledge from 111 different language editions of", "key_claims": [ "Large-scale structured knowledge can be extracted from semi-structured community resources.", "Linked Data enables interconnection across datasets." ], "ontology_relevance": "Prominent linked-data knowledge graph and ontology mapping example.", "ai_relevance": "Common resource for entity linking, QA, and KG research.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "commercial", "core", "dbpedia", "foundational", "knowledge-graph", "linked-data", "openalex" ], "triage_tier": "core", "triage_score": 111 }, { "id": "phase2-gutierrez-2025-hipporag2", "title": "From RAG to Memory: Non-Parametric Continual Learning for Large Language Models", "authors_or_org": "Bernal Jimenez Gutierrez; Yiheng Shu; Weijian Qi; Sizhe Zhou; Yu Su", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2502.14802", "doi_or_identifier": "arxiv:2502.14802", "venue_or_site": "ICML 2025 / PMLR", "abstract_or_summary": "Presents HippoRAG 2, extending graph-based RAG toward non-parametric continual learning with improved factual, sense-making, and associative memory.", "key_claims": [ "RAG can be reframed as non-parametric continual learning rather than only answer retrieval.", "Graph-augmented memory must preserve factual accuracy while improving associativity and sense-making.", "Continual AI memory requires organization and retrieval policies, not only embedding stores.", "RAG can be reframed as non-parametric memory rather than transient document lookup.", "Continual memory requires updates and retrieval consistency across sessions.", "Memory-oriented RAG raises stronger requirements for provenance, correction, and forgetting." ], "ontology_relevance": "Important for arguing that ontologies/KGs are memory infrastructure for AI agents.", "ai_relevance": "Recent peer-reviewed source on memory-oriented RAG and continual learning.", "palantir_relevance": "Useful conceptual comparator to Palantir's operational ontology as persistent enterprise memory.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "agent-memory", "continual-learning", "core", "graph-rag", "hipporag", "knowledge-graph", "memory", "non-parametric-memory", "rag" ], "triage_tier": "core", "triage_score": 110 }, { "id": "phase2-llms4om-2024", "title": "LLMs4OM: Matching Ontologies with Large Language Models", "authors_or_org": "Hamed Babaei Giglou; Jennifer D'Souza; Felix Engel; Soren Auer", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2404.10317", "doi_or_identifier": "arxiv:2404.10317", "venue_or_site": "arXiv / ESWC", "abstract_or_summary": "Evaluates large language models for ontology matching using retrieval and matching modules across concept, parent, and children representations over multiple OM datasets.", "key_claims": [ "LLMs can support ontology matching when prompts include structural context around concepts.", "Retrieval plus matching is a useful architecture for aligning heterogeneous ontologies.", "Ontology matching remains a key semantic interoperability task for multi-source AI systems.", "Retrieval-before-matching helps control LLM cost and candidate scope in ontology matching.", "Parent and child context can improve concept matching beyond label-only prompts.", "LLMs can outperform or match traditional systems in some complex matching cases." ], "ontology_relevance": "Key recent source for LLM-assisted ontology alignment and interoperability.", "ai_relevance": "Shows how LLMs can be integrated into schema and ontology matching workflows.", "palantir_relevance": "Relevant to enterprise ontology integration where multiple domain models must be reconciled.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "core", "eswc-2024", "interoperability", "llm", "ontology-matching", "retrieval", "schema-alignment", "semantic-integration" ], "triage_tier": "core", "triage_score": 110 }, { "id": "core-lewis-2020-rag", "title": "Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks", "authors_or_org": "Patrick Lewis; Ethan Perez; Aleksandra Piktus; Fabio Petroni; Vladimir Karpukhin; Naman Goyal; Heinrich Kuttler; Mike Lewis; Wen-tau Yih; Tim Rocktaschel; Sebastian Riedel; Douwe Kiela", "year": 2020, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2005.11401", "doi_or_identifier": "arxiv:2005.11401", "venue_or_site": "NeurIPS", "abstract_or_summary": "Introduces RAG models that combine parametric generation with non-parametric retrieval for knowledge-intensive NLP.", "key_claims": [ "Retrieval can ground generation in external knowledge sources.", "Hybrid parametric/non-parametric memory improves knowledge-intensive tasks.", "Retrieval can improve knowledge-intensive generation tasks.", "Separating parametric and non-parametric memory supports knowledge updates.", "Generated answers depend on retriever quality and evidence selection.", "External retrieval separates mutable knowledge from model parameters.", "Answer quality depends on retriever quality and evidence selection.", "RAG establishes the baseline that graph and ontology systems extend." ], "ontology_relevance": "Provides retrieval architecture that ontology-indexed chunks can improve.", "ai_relevance": "Foundational source for modern RAG systems.", "palantir_relevance": "Supports enterprise pattern of retrieving authoritative operational context before generation.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "core", "evidence", "external-memory", "grounding", "knowledge-intensive", "knowledge-intensive-nlp", "llm", "rag", "retrieval", "retrieval-augmented-generation" ], "triage_tier": "core", "triage_score": 110 }, { "id": "core-berners-lee-2001-semantic-web", "title": "The Semantic Web", "authors_or_org": "Tim Berners-Lee; James Hendler; Ora Lassila", "year": 2001, "source_type": "technical_article", "bucket": "technical", "url": "https://doi.org/10.1038/scientificamerican0501-34", "doi_or_identifier": "10.1038/scientificamerican0501-34", "venue_or_site": "Scientific American", "abstract_or_summary": "Landmark article presenting a vision of Web data with machine-interpretable meaning through agents, ontologies, and shared semantics.", "key_claims": [ "The Semantic Web aims to make Web content processable by machines through explicit semantics.", "Ontologies allow agents and services to interpret data across sources.", "The Web can be extended with structured meaning that software agents can process.", "Ontologies and metadata help machines interpret data from diverse sources.", "Machine-actionable semantics can support automated services and coordination." ], "ontology_relevance": "Popularized ontology as infrastructure for machine-understandable Web data.", "ai_relevance": "Early agent-oriented vision of semantic interoperability.", "palantir_relevance": "Relevant as a precursor to enterprise semantic layers that let software act over integrated data.", "quality_signal": "widely_cited", "retrieval_tags": [ "agents", "core", "linked-data", "machine-readable-data", "metadata", "semantic-web" ], "triage_tier": "core", "triage_score": 110 }, { "id": "oa-https-doi-org-10-3233-sw-160218", "title": "Knowledge graph refinement: A survey of approaches and evaluation methods", "authors_or_org": "Heiko Paulheim", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/SW-160218", "doi_or_identifier": "10.3233/sw-160218", "venue_or_site": "Semantic Web", "abstract_or_summary": "In the recent years, different Web knowledge graphs, both free and commercial, have been created. While Google coined the term “Knowledge Graph” in 2012, there are also a few openly available knowledge graphs, with DBpedia, YAGO, and Freebase being a", "key_claims": [ "KG refinement covers completion, error detection, type assertion, and relation prediction.", "Evaluation must distinguish different refinement tasks and data assumptions." ], "ontology_relevance": "Important for maintaining ontology-backed knowledge bases over time.", "ai_relevance": "Relevant to AI-assisted graph completion and quality control.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "core", "foundational", "knowledge-graph", "openalex", "quality", "refinement", "survey" ], "triage_tier": "core", "triage_score": 109 }, { "id": "core-bfo-iso-21838-2", "title": "Information technology - Top-level ontologies - Part 2: Basic Formal Ontology (BFO)", "authors_or_org": "ISO/IEC", "year": 2021, "source_type": "standard", "bucket": "technical", "url": "https://www.iso.org/standard/74572.html", "doi_or_identifier": "iso/iec 21838-2:2021", "venue_or_site": "ISO/IEC", "abstract_or_summary": "International standard specifying Basic Formal Ontology as a top-level ontology for interoperable domain ontologies.", "key_claims": [ "BFO supplies upper-level categories for domain ontology construction.", "A top-level ontology can improve cross-domain consistency and interoperability.", "BFO is standardized as a top-level ontology for broad interoperability.", "Top-level ontology standards can support consistent domain ontology development.", "BFO provides high-level categories for entities, continuants, occurrents, qualities, roles, and relations." ], "ontology_relevance": "Standardized upper ontology used in biomedical and industrial domains.", "ai_relevance": "Useful for shared semantic grounding in multi-domain AI systems.", "palantir_relevance": "Useful reference for evaluating enterprise object and process modeling against a formal upper ontology.", "quality_signal": "official_standard", "retrieval_tags": [ "bfo", "core", "interoperability", "iso-21838", "standard", "top-level-ontology", "upper-ontology" ], "triage_tier": "core", "triage_score": 108 }, { "id": "phase2-li-garijo-poveda-2025-llm-oe-review", "title": "Large Language Models for Ontology Engineering: A Systematic Literature Review", "authors_or_org": "J. Li; D. Garijo; M. Poveda-Villalon", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://www.semantic-web-journal.net/content/large-language-models-ontology-engineering-systematic-literature-review", "doi_or_identifier": null, "venue_or_site": "Semantic Web Journal preprint", "abstract_or_summary": "Systematic review of how large language models are being applied across ontology engineering tasks, including generation, refinement, alignment, evaluation, and human-in-the-loop workflows.", "key_claims": [ "LLMs are being applied across multiple ontology engineering tasks, but evaluation and reliability remain uneven.", "Ontology engineering with LLMs requires human oversight, task decomposition, and quality controls.", "The literature is shifting from isolated prompting experiments toward workflow-level OE assistance.", "LLMs are being used as ontology engineers, domain experts, and evaluators across the ontology lifecycle.", "Current studies lack stable task definitions, shared benchmarks, and reproducible evaluation protocols.", "Hybrid workflows with human expertise remain necessary for credible ontology engineering." ], "ontology_relevance": "High-value survey for the LLM-assisted ontology engineering research frontier.", "ai_relevance": "Maps LLM roles in generating, refining, validating, and aligning symbolic knowledge structures.", "palantir_relevance": "Provides academic context for comparing Palantir's ontology-assisted AI tooling with broader OE research.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "benchmarks", "core", "human-in-the-loop", "llm", "ontology-engineering", "ontology-governance", "survey", "systematic-review", "validation" ], "triage_tier": "core", "triage_score": 108 }, { "id": "core-w3c-owl2-overview", "title": "OWL 2 Web Ontology Language Document Overview", "authors_or_org": "W3C OWL Working Group", "year": 2012, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/owl2-overview", "doi_or_identifier": "w3c:rec-owl2-overview-20121211", "venue_or_site": "W3C", "abstract_or_summary": "W3C overview of OWL 2, a Semantic Web language for classes, properties, individuals, axioms, and reasoning profiles.", "key_claims": [ "OWL 2 supports formal ontology modeling with reasoning semantics.", "OWL profiles support different computational tradeoffs for applications.", "OWL 2 provides formal languages for expressing ontologies on the Web.", "OWL profiles trade expressivity for computational properties and implementation needs.", "OWL enables automated reasoning such as classification and consistency checking.", "OWL supports formal modeling of classes, properties, individuals, and logical axioms.", "OWL semantics enable automated reasoning over ontologies.", "Different OWL profiles balance expressivity and computational tractability." ], "ontology_relevance": "Authoritative standard for Web ontology representation and reasoning.", "ai_relevance": "Defines machine-checkable semantics useful for explainable and reasoning-capable AI.", "palantir_relevance": "Relevant where operational ontology needs class semantics, property restrictions, or formal interoperability.", "quality_signal": "official_standard", "retrieval_tags": [ "core", "description-logic", "knowledge-representation", "ontology", "ontology-language", "owl", "reasoning", "semantic-web", "standard", "standards", "w3c" ], "triage_tier": "core", "triage_score": 108 }, { "id": "core-baader-2003-description-logic-handbook", "title": "The Description Logic Handbook: Theory, Implementation, and Applications", "authors_or_org": "Franz Baader; Diego Calvanese; Deborah McGuinness; Daniele Nardi; Peter Patel-Schneider", "year": 2003, "source_type": "book", "bucket": "books", "url": "https://www.cambridge.org/core/books/description-logic-handbook", "doi_or_identifier": "isbn 9780521781763", "venue_or_site": "Cambridge University Press", "abstract_or_summary": "Reference handbook on description logics, the formal basis of many ontology languages including OWL.", "key_claims": [ "Description logics balance expressivity and decidable reasoning.", "Ontology languages can be grounded in formal semantics and automated inference." ], "ontology_relevance": "Formal reasoning foundation for ontology languages.", "ai_relevance": "Connects symbolic reasoning, knowledge representation, and tractable inference.", "palantir_relevance": "", "quality_signal": "scholarly_book", "retrieval_tags": [ "book", "core", "description-logic", "owl", "reasoning" ], "triage_tier": "core", "triage_score": 106 }, { "id": "core-guarino-1998-formal-ontology-information-systems", "title": "Formal Ontology in Information Systems", "authors_or_org": "Nicola Guarino", "year": 1998, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.5555/521720.521722", "doi_or_identifier": "fois 1998", "venue_or_site": "FOIS", "abstract_or_summary": "Positions formal ontology as a foundation for information systems, conceptual modeling, and shared meaning across systems.", "key_claims": [ "Formal ontology helps clarify conceptual distinctions in information systems.", "Ontology is useful when data structures need meaning-preserving integration." ], "ontology_relevance": "Foundational bridge between philosophical ontology and information-system modeling.", "ai_relevance": "Important for semantic interoperability and model grounding.", "palantir_relevance": "", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "conceptual-modeling", "core", "formal-ontology", "information-systems" ], "triage_tier": "core", "triage_score": 104 }, { "id": "phase2-he-2024-g-retriever", "title": "G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering", "authors_or_org": "Xiaoxin He; Yijun Tian; Yifei Sun; Nitesh V. Chawla; Thomas Laurent; Yann LeCun; Xavier Bresson; Bryan Hooi", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2402.07630", "doi_or_identifier": "arxiv:2402.07630", "venue_or_site": "NeurIPS 2024 / arXiv", "abstract_or_summary": "Develops a RAG framework for real-world textual graphs and graph question answering, including graph-relevant retrieval via Prize-Collecting Steiner Tree optimization.", "key_claims": [ "Graph question answering needs retrieval over graph structure, not only document chunks.", "Optimizing subgraph selection can reduce hallucination and fit graph evidence into the LLM context window.", "Textual graphs are a practical setting where KG reasoning and LLM generation meet.", "Graph QA needs retrieval over both topology and node text.", "Subgraph retrieval can reduce context size while preserving structural evidence.", "Text-only retrieval misses information encoded in graph neighborhoods." ], "ontology_relevance": "Useful for retrieving evidence from ontology-backed textual graphs.", "ai_relevance": "Strong peer-reviewed source for graph QA and RAG over structured graph evidence.", "palantir_relevance": "Comparable to querying Palantir object/link graphs through AI interfaces.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "core", "g-retriever", "graph-qa", "graph-rag", "hallucination", "knowledge-graph", "rag", "retrieval", "subgraph-retrieval", "textual-graphs" ], "triage_tier": "core", "triage_score": 104 }, { "id": "core-staab-studer-2009-handbook-ontologies", "title": "Handbook on Ontologies", "authors_or_org": "Steffen Staab; Rudi Studer", "year": 2009, "source_type": "book", "bucket": "books", "url": "https://doi.org/10.1007/978-3-540-92673-3", "doi_or_identifier": "10.1007/978-3-540-92673-3", "venue_or_site": "Springer", "abstract_or_summary": "Edited handbook covering ontology languages, engineering methods, ontology learning, matching, reasoning, and applications.", "key_claims": [ "Ontology engineering combines formal languages, methods, tools, and application practices.", "Ontology lifecycle includes modeling, alignment, learning, evaluation, and reuse.", "Ontology research spans formal foundations, engineering methods, languages, and applications.", "Evaluation, alignment, reasoning, and lifecycle management are central ontology concerns.", "Ontology systems require both conceptual rigor and implementation tooling." ], "ontology_relevance": "Broad reference for ontology engineering and applications.", "ai_relevance": "Maps the technical ecosystem that modern AI can reuse or automate.", "palantir_relevance": "Provides categories for analyzing enterprise ontology capabilities beyond product terminology.", "quality_signal": "widely_cited", "retrieval_tags": [ "book", "core", "handbook", "mapping", "matching", "ontology-engineering", "ontology-learning", "reasoning", "semantic-web" ], "triage_tier": "core", "triage_score": 104 }, { "id": "ont-ai-023", "title": "HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models", "authors_or_org": "Liang Wang; Nan Yang; Furu Wei; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2405.14831", "doi_or_identifier": "arxiv:2405.14831", "venue_or_site": "arXiv", "abstract_or_summary": "Presents a long-term memory retrieval framework for LLMs inspired by hippocampal indexing, using graph-based associations for multi-hop retrieval.", "key_claims": [ "LLM memory benefits from associative structures beyond flat vector search.", "Graph-like retrieval supports multi-hop and cross-document reasoning.", "Memory quality depends on extraction, indexing, and retrieval mechanisms.", "Knowledge graphs can act as associative memory structures for LLM retrieval.", "Personalized PageRank over graph memory can reduce cost and improve multi-hop retrieval compared with iterative retrieval.", "Agent memory should integrate new experiences without relying only on parametric retraining.", "Associative graph memory can retrieve evidence missed by flat vector search.", "Multi-hop questions benefit from paths across entities and propositions.", "Memory quality depends on extraction, indexing, and graph traversal design.", "Associative graph memory can improve multi-hop retrieval beyond flat RAG.", "Single-step graph retrieval can be cheaper and faster than some iterative retrieval methods.", "Long-term LLM memory needs mechanisms for integrating new knowledge without retraining." ], "ontology_relevance": "Relevant to ontology-backed agent memory and structured retrieval.", "ai_relevance": "Recent contribution to long-term memory for LLM systems.", "palantir_relevance": "Relevant to persistent operational memory over entities and events.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "agent-memory", "associative-memory", "core", "graph-retrieval", "hipporag", "knowledge-graph", "llm", "long-term-memory", "multi-hop", "multi-hop-qa", "neurips-2024", "pagerank", "rag" ], "triage_tier": "core", "triage_score": 104 }, { "id": "core-sowa-2000-knowledge-representation", "title": "Knowledge Representation: Logical, Philosophical, and Computational Foundations", "authors_or_org": "John F. Sowa", "year": 2000, "source_type": "book", "bucket": "books", "url": "https://www.jfsowa.com/krbook", "doi_or_identifier": "isbn 9780534949655", "venue_or_site": "Brooks/Cole", "abstract_or_summary": "Comprehensive treatment of logic, ontology, conceptual graphs, philosophy, and computation for knowledge representation.", "key_claims": [ "Knowledge representation sits between logic, philosophy, linguistics, and computation.", "Ontological categories influence how systems model and reason about the world.", "Knowledge representation requires choices about logic, ontology, language, and computation.", "No single representation captures every purpose; expressive power and computational tractability trade off.", "Conceptual structures mediate between natural language, databases, and reasoning systems." ], "ontology_relevance": "Rich bridge between philosophical ontology and computational modeling.", "ai_relevance": "Useful for grounding ontology in AI representation choices.", "palantir_relevance": "Helps compare enterprise ontology with classic KR layers connecting human language, data, and inference.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-foundations", "book", "conceptual-graphs", "core", "formal-semantics", "knowledge-representation", "logic", "ontology", "phase3" ], "triage_tier": "core", "triage_score": 104 }, { "id": "core-uschold-gruninger-1996-principles-methods", "title": "Ontologies: Principles, Methods and Applications", "authors_or_org": "Mike Uschold; Michael Gruninger", "year": 1996, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1017/S0269888900007797", "doi_or_identifier": "10.1017/s0269888900007797", "venue_or_site": "Knowledge Engineering Review", "abstract_or_summary": "Survey and methodological account of ontology construction, ontology uses, and engineering practices.", "key_claims": [ "Ontologies support communication, interoperability, and systems engineering.", "Ontology construction benefits from explicit scope, competency questions, reuse, and evaluation.", "Ontology development should be driven by purpose and intended use.", "Competency questions help specify what an ontology must support.", "Reuse and evaluation are part of ontology engineering, not optional afterthoughts." ], "ontology_relevance": "Classic ontology engineering methodology source.", "ai_relevance": "Gives practical engineering steps for AI knowledge models.", "palantir_relevance": "Useful for assessing whether an operational ontology supports concrete business questions and actions.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "competency-questions", "core", "evaluation", "methodology", "ontology-engineering" ], "triage_tier": "core", "triage_score": 104 }, { "id": "phase2-gangemi-2005-ontology-design-patterns", "title": "Ontology Design Patterns for Semantic Web Content", "authors_or_org": "Aldo Gangemi", "year": 2005, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/11574620_21", "doi_or_identifier": "10.1007/11574620_21", "venue_or_site": "International Semantic Web Conference", "abstract_or_summary": "Introduces ontology design patterns as reusable modeling solutions for recurring ontology engineering problems in Semantic Web content.", "key_claims": [ "Ontology design patterns improve reuse, interoperability, and modeling quality.", "Recurring modeling problems should be solved through shared patterns rather than ad hoc class hierarchies.", "LLM-assisted ontology construction can benefit from pattern libraries as constraints and examples." ], "ontology_relevance": "Core source for reusable ontology modeling patterns and engineering discipline.", "ai_relevance": "Relevant to using pattern libraries to constrain LLM-generated schemas and relations.", "palantir_relevance": "Useful comparator to Palantir ontology best practices and object/action modeling conventions.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "core", "ontology-design-patterns", "ontology-engineering", "reuse", "semantic-web", "validation" ], "triage_tier": "core", "triage_score": 104 }, { "id": "iso_common_logic_2018", "title": "ISO/IEC 24707:2018 Information Technology - Common Logic", "authors_or_org": "International Organization for Standardization and International Electrotechnical Commission", "year": 2018, "source_type": "docs", "bucket": "technical", "url": "https://www.iso.org/obp/ui/en", "doi_or_identifier": "iso/iec 24707:2018", "venue_or_site": "ISO", "abstract_or_summary": "International standard for a family of first-order logic-based languages intended for information exchange and transmission of logical content.", "key_claims": [ "Common Logic standardizes exchange of logic-based information.", "It supports expressive axiomatization beyond lightweight taxonomy or graph structures.", "Logic interchange requires precise syntax and semantics.", "Common Logic supplies a standard interchange framework for logic-based knowledge representation.", "Ontology interchange requires attention to syntax, semantics, and expressivity boundaries.", "Logic standards remain relevant when AI systems need formal commitments rather than loose labels." ], "ontology_relevance": "Important standard for highly formal ontologies and interoperation among logic-based systems.", "ai_relevance": "Relevant when AI systems need stronger axioms, theorem proving, or formal verification beyond OWL profiles.", "palantir_relevance": "Indirectly relevant to cases where enterprise semantics require first-order constraints rather than graph navigation alone.", "quality_signal": "official_standard", "retrieval_tags": [ "common-logic", "core", "first-order-logic", "formal-ontology", "interoperability", "iso", "knowledge-representation", "logic", "standard" ], "triage_tier": "core", "triage_score": 102 }, { "id": "ont-ai-022", "title": "LightRAG: Simple and Fast Retrieval-Augmented Generation", "authors_or_org": "Zirui Guo; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2410.05779", "doi_or_identifier": "arxiv:2410.05779", "venue_or_site": "arXiv", "abstract_or_summary": "Proposes a lightweight graph-based RAG approach that combines local and global retrieval over graph structures for efficient generation.", "key_claims": [ "Graph-enhanced retrieval can be made more efficient than heavier graph pipelines.", "Combining local and global context can improve answer quality.", "RAG systems must balance retrieval depth, latency, and graph maintenance cost.", "Flat vector retrieval misses interdependencies that graph structures can expose.", "Dual-level retrieval can combine entity-level detail with higher-level knowledge discovery.", "Incremental graph updates are important for dynamic knowledge bases.", "Graph-enhanced retrieval can be simplified for lower latency and easier deployment.", "Combining local entity detail with global conceptual context improves retrieval coverage.", "Graph RAG systems must trade retrieval depth against update and runtime cost.", "Flat vector retrieval can miss dependencies that graph structure captures.", "Dual-level retrieval can combine local entity context with higher-level knowledge discovery.", "Incremental updates are important for RAG systems over changing domains." ], "ontology_relevance": "Shows a practical direction for graph-structured retrieval over domain knowledge.", "ai_relevance": "Recent RAG architecture for LLM applications.", "palantir_relevance": "Relevant to low-latency operational retrieval over ontology-linked data.", "quality_signal": "secondary_source", "retrieval_tags": [ "core", "efficient-rag", "graph-rag", "graphrag", "incremental-update", "knowledge-graph", "lightrag", "llm", "local-global-retrieval", "rag", "retrieval", "structured-retrieval", "vector-graph-hybrid" ], "triage_tier": "core", "triage_score": 102 }, { "id": "core-besold-2017-neural-symbolic-survey", "title": "Neural-Symbolic Learning and Reasoning: A Survey and Interpretation", "authors_or_org": "Tarek R. Besold; Artur d'Avila Garcez; Sebastian Bader; Howard Bowman; Pedro Domingos; Pascal Hitzler; Kai-Uwe Kuhnberger; Luis C. Lamb; Daniel Lowd; Priscila Machado Vieira Lima; Leo de Penning; Gadi Pinkas; Hoifung Poon; Gerson Zaverucha", "year": 2017, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/1711.03902", "doi_or_identifier": "arxiv:1711.03902", "venue_or_site": "arXiv", "abstract_or_summary": "Survey and interpretation of neural-symbolic methods that combine learning with logic, representation, and reasoning.", "key_claims": [ "Neural-symbolic systems seek to combine robust learning with explicit reasoning.", "Key issues include representation, extraction, integration, and explainability.", "Neural and symbolic AI offer complementary strengths.", "Neural-symbolic integration can support reasoning, learning from examples, and interpretability.", "Scaling and representation translation remain major challenges." ], "ontology_relevance": "Places ontologies inside hybrid symbolic/neural AI programs.", "ai_relevance": "Important bridge from KR/ontology to modern neural AI.", "palantir_relevance": "Supports the conceptual basis for ontology plus LLM operational AI.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "core", "logic", "neuro-symbolic", "neuro-symbolic-ai", "reasoning", "survey" ], "triage_tier": "core", "triage_score": 102 }, { "id": "core-garcez-lamb-2023-neurosymbolic-third-wave", "title": "Neurosymbolic AI: The 3rd Wave", "authors_or_org": "Artur d'Avila Garcez; Luis C. Lamb", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.artint.2023.103960", "doi_or_identifier": "10.1016/j.artint.2023.103960", "venue_or_site": "Artificial Intelligence", "abstract_or_summary": "Position paper arguing for neurosymbolic AI as a third wave that integrates neural learning and symbolic reasoning.", "key_claims": [ "Neurosymbolic AI aims to combine learning, reasoning, and explanation.", "Future AI systems need integration across statistical and symbolic representations." ], "ontology_relevance": "Justifies ontology as part of the symbolic substrate for hybrid AI.", "ai_relevance": "Current high-level framing for neural-symbolic integration.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "ai", "core", "neuro-symbolic", "reasoning" ], "triage_tier": "core", "triage_score": 102 }, { "id": "core-wong-2012-ontology-learning-text", "title": "Ontology Learning from Text: A Look Back and into the Future", "authors_or_org": "Wilson Wong; Wei Liu; Mohammed Bennamoun", "year": 2012, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1145/2333112.2333115", "doi_or_identifier": "10.1145/2333112.2333115", "venue_or_site": "ACM Computing Surveys", "abstract_or_summary": "Survey of ontology learning from text, covering term extraction, concept formation, relation extraction, taxonomy induction, and evaluation.", "key_claims": [ "Ontology learning automates parts of ontology construction from textual corpora.", "Key challenges include evaluation, relation extraction, and moving from terms to conceptual structure.", "Ontology learning remains difficult because semantic interpretation and evaluation are hard.", "Different ontology components require different extraction methods.", "Future systems need better integration of statistical, linguistic, and knowledge-based methods." ], "ontology_relevance": "Core survey for automating ontology construction.", "ai_relevance": "Directly relevant to LLM-assisted ontology extraction and maintenance.", "palantir_relevance": "Useful background for why automated ontology construction is valuable but insufficient by itself.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "core", "evaluation", "ontology-learning", "survey", "text-mining" ], "triage_tier": "core", "triage_score": 102 }, { "id": "core-shvaiko-euzenat-2013-matching-state-art", "title": "Ontology Matching: State of the Art and Future Challenges", "authors_or_org": "Pavel Shvaiko; Jerome Euzenat", "year": 2013, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1109/TKDE.2011.253", "doi_or_identifier": "10.1109/tkde.2011.253", "venue_or_site": "IEEE Transactions on Knowledge and Data Engineering", "abstract_or_summary": "Survey of ontology matching approaches, evaluations, systems, and future challenges.", "key_claims": [ "Ontology matching combines terminological, structural, semantic, and instance-based evidence.", "Open challenges include scalability, uncertainty, user interaction, and evaluation." ], "ontology_relevance": "High-value survey for alignment methods and limitations.", "ai_relevance": "Useful for LLM-assisted schema/ontology alignment and data integration.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "alignment", "core", "evaluation", "interoperability", "ontology-matching", "survey" ], "triage_tier": "core", "triage_score": 102 }, { "id": "core-gruber-1993-portable-ontology", "title": "A Translation Approach to Portable Ontology Specifications", "authors_or_org": "Thomas R. Gruber", "year": 1993, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1006/knac.1993.1008", "doi_or_identifier": "10.1006/knac.1993.1008", "venue_or_site": "Knowledge Acquisition", "abstract_or_summary": "Foundational paper defining an ontology as an explicit specification of a conceptualization and framing ontologies as reusable knowledge-sharing artifacts.", "key_claims": [ "Ontologies enable knowledge sharing by making conceptual commitments explicit.", "A portable ontology specification can support translation across representation systems.", "Ontologies support sharing and reuse of knowledge among agents and systems.", "An ontology can be treated as an explicit specification of a conceptualization.", "Portability requires separating conceptual commitments from implementation details." ], "ontology_relevance": "Canonical computer-science definition of ontology for knowledge sharing.", "ai_relevance": "Grounds the knowledge-representation lineage of ontology-based AI.", "palantir_relevance": "Supports the idea of an ontology layer that decouples operational semantics from underlying source systems.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "conceptualization", "core", "formal-ontology", "gruber", "knowledge-sharing", "ontology-definition", "portability" ], "triage_tier": "core", "triage_score": 98 }, { "id": "phase2-kommineni-2024-llm-ontology-kg-construction", "title": "From human experts to machines: An LLM supported approach to ontology and knowledge graph construction", "authors_or_org": "Vamsi Krishna Kommineni; Birgitta Konig-Ries; Sheeba Samuel", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2403.08345", "doi_or_identifier": "arxiv:2403.08345", "venue_or_site": "arXiv", "abstract_or_summary": "Presents a semi-automatic pipeline using LLMs to formulate competency questions, develop an ontology TBox, construct a KG, and evaluate results with reduced human involvement.", "key_claims": [ "Competency questions can be used as an organizing bridge between human requirements and LLM-supported ontology construction.", "LLM-generated KGs still need evaluation and human-in-the-loop quality checks.", "A pipeline view is more useful than single-prompt ontology generation for practical knowledge engineering.", "LLMs can reduce expert effort across competency question creation, ontology modeling, and KG construction.", "LLM-as-judge evaluation is useful for iteration but cannot replace expert validation.", "Human-in-the-loop checks remain necessary for automatically generated KGs." ], "ontology_relevance": "Directly relevant to article sections on automating ontology construction and evaluation.", "ai_relevance": "Shows a concrete LLM pipeline for TBox/ABox-style knowledge graph construction.", "palantir_relevance": "Useful as an academic counterpoint to Palantir's productized ontology-building tools.", "quality_signal": "preprint", "retrieval_tags": [ "competency-questions", "core", "evaluation", "human-in-the-loop", "knowledge-graph-construction", "llm", "ontology-construction", "ontology-engineering" ], "triage_tier": "core", "triage_score": 98 }, { "id": "core-gene-ontology-2000", "title": "Gene Ontology: Tool for the Unification of Biology", "authors_or_org": "The Gene Ontology Consortium", "year": 2000, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/75556", "doi_or_identifier": "10.1038/75556", "venue_or_site": "Nature Genetics", "abstract_or_summary": "Introduces the Gene Ontology as a controlled vocabulary for molecular function, biological process, and cellular component across organisms.", "key_claims": [ "A shared ontology enables consistent annotation of gene products across databases and species.", "Community-maintained ontologies can become scientific infrastructure.", "Shared biological vocabularies enable cross-database annotation and comparison.", "Ontology can unify data across species and research communities.", "Community curation and standard terms are central to scientific data integration." ], "ontology_relevance": "Canonical applied ontology success case.", "ai_relevance": "Shows ontology as high-value training, retrieval, and annotation infrastructure in biomedicine.", "palantir_relevance": "Relevant as an example of domain ontology becoming operational infrastructure for analytics.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "annotation", "biology", "biomedical-ontology", "controlled-vocabulary", "core", "data-integration", "gene-ontology", "scientific-infrastructure" ], "triage_tier": "core", "triage_score": 98 }, { "id": "core-brachman-levesque-2004-kr", "title": "Knowledge Representation and Reasoning", "authors_or_org": "Ronald J. Brachman; Hector J. Levesque", "year": 2004, "source_type": "book", "bucket": "books", "url": "https://shop.elsevier.com/books/knowledge-representation-and-reasoning/brachman/978-1-55860-932-7", "doi_or_identifier": "isbn 9781558609327", "venue_or_site": "Morgan Kaufmann", "abstract_or_summary": "Textbook covering logic, frames, semantic networks, description logics, reasoning, and KR principles.", "key_claims": [ "Knowledge representation is the study of how to encode information so systems can reason with it.", "Different KR formalisms make different tradeoffs between expressiveness and reasoning cost.", "Representation choices determine what can be inferred and how efficiently.", "Knowledge representation must balance expressivity, tractability, and usefulness.", "Reasoning systems require formal semantics, not just data structures." ], "ontology_relevance": "Places ontology inside the broader KR tradition.", "ai_relevance": "Essential background for AI systems that combine data, symbols, and inference.", "palantir_relevance": "Relevant to operational AI architectures where ontology-backed data must support queries, inference, and actions.", "quality_signal": "widely_cited", "retrieval_tags": [ "book", "core", "expressivity", "knowledge-representation", "reasoning", "symbolic-ai" ], "triage_tier": "core", "triage_score": 98 }, { "id": "phase2-mcp-spec-2025", "title": "Model Context Protocol Specification", "authors_or_org": "Model Context Protocol project", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://modelcontextprotocol.io/specification/2025-06-18", "doi_or_identifier": "mcp specification 2025-06-18", "venue_or_site": "modelcontextprotocol.io", "abstract_or_summary": "Official MCP specification defining a protocol for connecting LLM applications with external data sources and tools using standardized client-server interactions.", "key_claims": [ "MCP standardizes how LLM applications integrate with external context, resources, and tools.", "Tool exposure requires schemas and protocol-level affordances that agents can discover and invoke.", "MCP turns tool access into an interoperability layer for agentic AI systems.", "Agent systems need a standard interface for tools and contextual resources.", "Resources and tools can expose structured enterprise context rather than relying only on text prompts.", "Protocol-level boundaries help separate model reasoning from external system capabilities." ], "ontology_relevance": "Relevant to exposing ontology objects, queries, and actions as structured agent tools.", "ai_relevance": "Core protocol source for agent tool-use and context integration.", "palantir_relevance": "Directly underpins Palantir Ontology MCP's public architecture claims.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "context-protocol", "core", "interoperability", "mcp", "ontology-mcp", "semantic-layer", "tool-use" ], "triage_tier": "core", "triage_score": 98 }, { "id": "core-noy-mcguinness-2001-ontology-101", "title": "Ontology Development 101: A Guide to Creating Your First Ontology", "authors_or_org": "Natalya F. Noy; Deborah L. McGuinness", "year": 2001, "source_type": "technical_article", "bucket": "technical", "url": "https://protege.stanford.edu/publications/ontology_development/ontology101.pdf", "doi_or_identifier": "stanford ksl technical report ksl-01-05", "venue_or_site": "Stanford Knowledge Systems Laboratory", "abstract_or_summary": "Practical guide to defining classes, slots, facets, instances, and iterative ontology development using Protege-style modeling.", "key_claims": [ "Ontology development is iterative and should begin with scope and competency questions.", "Classes, relations, properties, constraints, and instances should be modeled deliberately.", "Ontology construction should begin with domain and scope questions.", "Existing ontologies should be considered before building from scratch.", "Classes, properties, constraints, and instances should be modeled iteratively." ], "ontology_relevance": "Highly cited practical entry point for ontology construction.", "ai_relevance": "Useful for translating research concepts into a working knowledge base schema.", "palantir_relevance": "Relevant as a pragmatic checklist for creating operational object models and relationships.", "quality_signal": "widely_cited", "retrieval_tags": [ "classes", "core", "ontology-development", "ontology-engineering", "properties", "protege", "slots", "tutorial" ], "triage_tier": "core", "triage_score": 98 }, { "id": "core-cimiano-2006-ontology-learning-population", "title": "Ontology Learning and Population from Text: Algorithms, Evaluation and Applications", "authors_or_org": "Philipp Cimiano", "year": 2006, "source_type": "book", "bucket": "books", "url": "https://link.springer.com/book/10.1007/978-0-387-39252-3", "doi_or_identifier": "10.1007/978-0-387-39252-3", "venue_or_site": "Springer", "abstract_or_summary": "Book on algorithms and evaluation for learning and populating ontologies from text.", "key_claims": [ "Ontology learning includes term, synonym, concept, taxonomy, relation, and axiom acquisition.", "Text-derived ontologies require evaluation against both linguistic evidence and conceptual adequacy.", "Ontology learning includes both schema acquisition and instance population.", "Text-derived ontologies need evaluation against expert knowledge and downstream use.", "Linguistic and statistical evidence are complementary for concept and relation discovery." ], "ontology_relevance": "Detailed reference for ontology learning pipelines.", "ai_relevance": "Pre-LLM foundation for automated semantic extraction.", "palantir_relevance": "Relevant to bootstrapping enterprise object models from documents and operational text.", "quality_signal": "widely_cited", "retrieval_tags": [ "book", "core", "nlp", "ontology-learning", "ontology-population", "population", "text", "text-mining" ], "triage_tier": "core", "triage_score": 98 }, { "id": "phase2-rigor-2025-rag-ontologies-rdb", "title": "Retrieval-Augmented Generation of Ontologies from Relational Databases", "authors_or_org": "Mojtaba Nayyeri; Athish A. Yogi; Nadeen Fathallah; Ratan Bahadur Thapa; Hans-Michael Tautenhahn; Anton Schnurpel; Steffen Staab", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2506.01232", "doi_or_identifier": "arxiv:2506.01232", "venue_or_site": "arXiv", "abstract_or_summary": "Presents RIGOR, a retrieval-augmented iterative approach for generating OWL ontologies from relational database schemas, documentation, domain ontologies, and a growing core ontology.", "key_claims": [ "RAG can support ontology generation by retrieving schema documentation, domain ontologies, and prior ontology fragments.", "Iterative delta ontology generation with judge-LLM refinement is more controllable than one-shot ontology generation.", "Foreign-key structure can guide the order of ontology construction from relational databases.", "Relational schemas can be converted into richer OWL ontologies through iterative table-by-table RAG workflows.", "Foreign-key structure provides a useful construction order for ontology generation from databases.", "Judge-LLM refinement and provenance-tagged delta fragments improve control over generated ontology changes." ], "ontology_relevance": "Direct bridge between database schemas, RAG, ontology generation, and quality evaluation.", "ai_relevance": "Concrete example of retrieval-augmented LLMs generating machine-readable ontologies.", "palantir_relevance": "Comparable to building operational ontologies from enterprise datasets and schemas.", "quality_signal": "preprint", "retrieval_tags": [ "core", "ontology-generation", "owl", "rag", "relational-database", "rigor", "schema", "semantic-layer", "validation" ], "triage_tier": "core", "triage_score": 98 }, { "id": "core-nickel-2016-relational-ml-kg", "title": "A Review of Relational Machine Learning for Knowledge Graphs", "authors_or_org": "Maximilian Nickel; Kevin Murphy; Volker Tresp; Evgeniy Gabrilovich", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://ieeexplore.ieee.org/document/7358050", "doi_or_identifier": "10.1109/jproc.2015.2483592", "venue_or_site": "Proceedings of the IEEE", "abstract_or_summary": "Review of statistical relational learning and representation learning methods for knowledge graphs.", "key_claims": [ "Knowledge graph embeddings support prediction over relational data.", "Relational learning complements symbolic KG semantics with statistical generalization.", "Relational structure is central to many AI tasks involving entities and links.", "Latent representations can infer missing relations in KGs.", "Combining symbolic graph structure with statistical learning is a recurring AI theme." ], "ontology_relevance": "Shows how structured ontological relations become learnable graph features.", "ai_relevance": "Foundation for KG embedding and link prediction approaches.", "palantir_relevance": "Relevant to predictive modeling over enterprise object-relation graphs.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "core", "embedding", "embeddings", "kg-completion", "knowledge-graph", "relational-learning" ], "triage_tier": "core", "triage_score": 96 }, { "id": "ont-ai-024", "title": "KAG: Boosting LLMs in Professional Domains via Knowledge Augmented Generation", "authors_or_org": "Mingyang Liang; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2409.13731", "doi_or_identifier": "arxiv:2409.13731", "venue_or_site": "arXiv", "abstract_or_summary": "Describes knowledge-augmented generation for professional domains, emphasizing structured knowledge and reasoning over domain-specific sources.", "key_claims": [ "Professional-domain LLM systems need structured knowledge and reasoning beyond generic RAG.", "Knowledge organization and retrieval strategy materially affect answer reliability.", "Domain schemas and knowledge bases can improve controllability.", "Professional-domain RAG needs logic, temporal relations, numerical constraints, and expert rules beyond vector similarity.", "Mutual indexing between knowledge graphs and text chunks supports both symbolic reasoning and evidence grounding.", "Hybrid reasoning engines can combine semantic reasoning, graph retrieval, and generation.", "Professional domains require structured knowledge and reasoning beyond generic vector RAG.", "Knowledge organization materially affects answer reliability and controllability.", "Hybrid retrieval should connect chunks, entities, relations, and logical reasoning.", "Professional RAG needs knowledge logic, temporal relations, expert rules, and graph reasoning beyond vector similarity.", "Mutual indexing between chunks and graph structure can improve traceability and retrieval.", "Domain QA systems benefit from combining knowledge reasoning, LLM reasoning, and mathematical logic." ], "ontology_relevance": "Supports the need for ontology-like knowledge organization in expert domains.", "ai_relevance": "Recent RAG variant focused on domain reliability.", "palantir_relevance": "Relevant to enterprise AI where domain objects and workflows matter.", "quality_signal": "secondary_source", "retrieval_tags": [ "core", "enterprise-ai", "graph-retrieval", "graphrag", "hybrid-retrieval", "kag", "knowledge-augmented-generation", "knowledge-graph", "llm", "logical-form", "openspg", "professional-domain", "rag", "semantic-layer", "semantic-reasoning" ], "triage_tier": "core", "triage_score": 96 }, { "id": "core-euzenat-shvaiko-2007-ontology-matching-book", "title": "Ontology Matching", "authors_or_org": "Jerome Euzenat; Pavel Shvaiko", "year": 2007, "source_type": "book", "bucket": "books", "url": "https://link.springer.com/book/10.1007/978-3-642-38721-0", "doi_or_identifier": "10.1007/978-3-642-38721-0", "venue_or_site": "Springer", "abstract_or_summary": "Comprehensive treatment of ontology matching, alignment methods, evaluation, and applications.", "key_claims": [ "Ontology matching identifies correspondences between semantically related entities across ontologies.", "Alignment is necessary for interoperability across independently built models." ], "ontology_relevance": "Core reference for ontology alignment and interoperability.", "ai_relevance": "Important for integrating multi-source knowledge in AI systems.", "palantir_relevance": "", "quality_signal": "scholarly_book", "retrieval_tags": [ "alignment", "book", "core", "interoperability", "ontology-matching" ], "triage_tier": "core", "triage_score": 94 }, { "id": "phase2-agent-om-2023", "title": "Agent-OM: Leveraging LLM Agents for Ontology Matching", "authors_or_org": "Zhangcheng Qiang; Weiqing Wang; Kerry Taylor", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2312.00326", "doi_or_identifier": "arxiv:2312.00326", "venue_or_site": "arXiv / PVLDB version available", "abstract_or_summary": "Introduces an agent-powered ontology matching framework with Siamese agents, retrieval, matching, and memory/tool components, evaluated on OAEI tracks.", "key_claims": [ "LLM agents can decompose ontology matching into retrieval, comparison, memory, and tool-use steps.", "Agentic ontology matching is especially promising for complex or few-shot matching tasks.", "Tool boundaries and memory design become part of semantic interoperability systems.", "LLM agents can structure ontology matching as tool-using retrieval and match decisions.", "Agentic matching improves complex and few-shot tasks compared with simpler LLM prompting.", "Tool design and task decomposition matter as much as the base LLM." ], "ontology_relevance": "Connects ontology matching to agent architectures and tool-assisted reasoning.", "ai_relevance": "Useful for the article's argument that ontology and agent tooling are converging.", "palantir_relevance": "Strong comparator to Palantir's Ontology MCP and controlled tool exposure for agents.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "agent-memory", "core", "llm-agent", "llm-agents", "oaei", "ontology-matching", "schema-alignment", "semantic-interoperability", "tool-use", "vldb-2025" ], "triage_tier": "core", "triage_score": 92 }, { "id": "pal-doc-aip-evals-ontology-edits-2026", "title": "AIP Evals: Evaluation functions and Ontology edits", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/aip-evals/ontology-edits", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Documentation on evaluating AIP Logic functions that create, edit, or delete ontology objects, including use of Ontology simulations so real ontology state remains unchanged during tests.", "key_claims": [ "Logic functions that involve ontology edits are executed in an Ontology simulation during evaluation.", "Created objects exist only in the simulated Ontology during tests.", "Custom TypeScript evaluation functions can verify edit outcomes." ], "ontology_relevance": "Shows simulation/testing mechanisms for write-capable ontology workflows.", "ai_relevance": "Important source for AI governance and validation of agentic edits.", "palantir_relevance": "Concrete evidence that AIP includes test mechanisms for ontology writeback behavior.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-evals", "governance", "ontology-edits", "palantir", "simulation", "testing" ], "triage_tier": "core", "triage_score": 92 }, { "id": "pal-doc-aip-observability-2026", "title": "AIP observability", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/aip/aip-observability", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing observability for AIP and Ontology workflow executions, including metrics, execution history, distributed tracing, logs, and log search.", "key_claims": [ "AIP observability covers AIP and Ontology workflow executions.", "Capabilities include metrics, execution history, distributed tracing, logging, and log search.", "The feature is part of platform-wide observability.", "The feature is part of platform-wide operational observability." ], "ontology_relevance": "Connects ontology-mediated execution to operational monitoring.", "ai_relevance": "Provides evidence for runtime governance and debugging of AI workflows.", "palantir_relevance": "Official source for AIP/Ontology observability claims.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-observability", "governance", "logs", "metrics", "palantir", "tracing" ], "triage_tier": "core", "triage_score": 92 }, { "id": "core-lenat-1995-cyc-large-knowledge-base", "title": "Cyc: A Midterm Report", "authors_or_org": "Douglas B. Lenat", "year": 1995, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1609/aimag.v11i3.842", "doi_or_identifier": "10.1609/aimag.v11i3.842", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "Reports on the Cyc project, a long-running attempt to encode common-sense knowledge in a large symbolic knowledge base.", "key_claims": [ "Large-scale common-sense knowledge requires extensive explicit representation.", "Symbolic knowledge bases can support reasoning but face acquisition and maintenance challenges.", "Common-sense AI requires large amounts of explicit background knowledge.", "Hand-built knowledge bases face scale, representation, and inference challenges.", "Microtheories can help organize context-dependent knowledge." ], "ontology_relevance": "Historical example of large explicit ontology/knowledge-base construction.", "ai_relevance": "Important contrast to data-driven and LLM-based AI approaches.", "palantir_relevance": "Relevant as background for large enterprise knowledge models and context-specific reasoning.", "quality_signal": "peer_reviewed_seminal", "retrieval_tags": [ "common-sense", "commonsense", "core", "cyc", "knowledge-base", "microtheories", "symbolic-ai" ], "triage_tier": "core", "triage_score": 92 }, { "id": "ont-ai-004", "title": "LLMs4OL: Large Language Models for Ontology Learning", "authors_or_org": "Hamed Babaei Giglou; Jennifer D'Souza; Sören Auer", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2307.16648", "doi_or_identifier": "arxiv:2307.16648", "venue_or_site": "arXiv", "abstract_or_summary": "Explores the use of large language models for ontology learning tasks, including term typing and taxonomy construction.", "key_claims": [ "LLMs can assist ontology learning tasks by exploiting broad language knowledge.", "LLM outputs still require validation because generated ontology elements may be inconsistent or unsupported.", "Ontology learning benchmarks need adaptation for generative models.", "LLMs can be evaluated on concrete ontology learning subtasks rather than vague ontology generation claims.", "Term typing, taxonomy discovery, and relation extraction provide a practical task decomposition for LLM-assisted ontology learning.", "Domain and genre variation remains a central challenge for ontology learning with LLMs.", "LLMs can assist ontology learning by proposing classes, terms, and taxonomic relations.", "Generated ontology elements may be inconsistent or unsupported without validation.", "Generative models require ontology-specific evaluation tasks and benchmarks.", "LLMs can perform several ontology learning subtasks without task-specific training.", "Performance varies by ontology task and knowledge domain, so a single LLM score is not enough.", "Ontology learning benchmarks need to account for generative outputs and semantic validity." ], "ontology_relevance": "Directly addresses LLM-assisted ontology creation.", "ai_relevance": "Treats LLMs as ontology engineering assistants rather than only text generators.", "palantir_relevance": "Relevant to semi-automated expansion of operational ontology classes and relations.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "benchmark", "core", "iswc-2023", "llm", "llms4ol", "ontology-engineering", "ontology-learning", "relation-extraction", "schema-extraction", "taxonomy", "taxonomy-discovery" ], "triage_tier": "core", "triage_score": 92 }, { "id": "core-niles-pease-2001-sumo", "title": "Towards a Standard Upper Ontology", "authors_or_org": "Ian Niles; Adam Pease", "year": 2001, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/505168.505170", "doi_or_identifier": "10.1145/505168.505170", "venue_or_site": "FOIS / IEEE", "abstract_or_summary": "Presents SUMO as a proposed upper ontology to support interoperability and general-purpose knowledge representation.", "key_claims": [ "A standard upper ontology can provide shared high-level categories across systems.", "SUMO aims to support broad knowledge engineering reuse.", "A standard upper ontology can provide shared high-level concepts for domain ontologies.", "SUMO integrates existing ontological resources and formal axiomatization.", "Upper ontology can improve semantic interoperability among systems." ], "ontology_relevance": "Important upper ontology project alongside BFO and DOLCE.", "ai_relevance": "Relevant to common-sense and cross-domain reasoning.", "palantir_relevance": "Useful comparison point for broad enterprise upper-model design.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "core", "formal-ontology", "interoperability", "standardization", "sumo", "upper-ontology" ], "triage_tier": "core", "triage_score": 92 }, { "id": "core-masolo-2003-dolce", "title": "WonderWeb Deliverable D18: Ontology Library (final)", "authors_or_org": "Claudio Masolo; Stefano Borgo; Aldo Gangemi; Nicola Guarino; Alessandro Oltramari", "year": 2003, "source_type": "technical_report", "bucket": "academic", "url": "http://www.loa.istc.cnr.it/old/DOLCE.html", "doi_or_identifier": "wonderweb d18", "venue_or_site": "Laboratory for Applied Ontology", "abstract_or_summary": "Introduces DOLCE and related foundational ontologies for cognitive and linguistic engineering.", "key_claims": [ "DOLCE models categories such as endurants, perdurants, qualities, and social objects.", "Foundational ontologies can clarify domain modeling choices." ], "ontology_relevance": "Major alternative upper ontology to BFO, strong in conceptual modeling.", "ai_relevance": "Helps AI systems distinguish objects, events, qualities, roles, and social constructs.", "palantir_relevance": "", "quality_signal": "widely_cited", "retrieval_tags": [ "core", "dolce", "foundational-ontology", "upper-ontology" ], "triage_tier": "core", "triage_score": 92 }, { "id": "pal-sec-2025-10k-2026", "title": "Palantir Technologies Inc. Form 10-K for fiscal year ended December 31, 2025", "authors_or_org": "Palantir Technologies Inc.", "year": 2026, "source_type": "other", "bucket": "palantir", "url": "https://www.sec.gov/Archives/edgar/data/1321655/000132165526000011/pltr-20251231.htm", "doi_or_identifier": "sec accession 0001321655-26-000011", "venue_or_site": "U.S. Securities and Exchange Commission EDGAR", "abstract_or_summary": "Annual report filed February 17, 2026. Describes Palantir's four principal platforms, including Foundry, AIP, Apollo, and Gotham; provides customer, revenue, risk, and business-strategy disclosures.", "key_claims": [ "Foundry provides data management, logic authoring, systemic mapping through Ontology, analytics, and workflow development.", "AIP provides secure LLM connectivity, agent/automation development tooling, AI-enabled applications, and evaluation frameworks.", "Palantir reported 954 customers as of December 31, 2025 and total 2025 revenue of about $4.475 billion.", "Palantir describes its platforms and business strategy in primary investor-facing terms.", "The filing is useful for distinguishing product architecture claims from business growth and risk disclosures.", "Risk factors provide counterweight to marketing claims about deployment, customer dependence, regulation, and competition." ], "ontology_relevance": "Primary business source confirming Ontology as a named capability of Foundry.", "ai_relevance": "Primary business source for AIP's product scope and production AI governance claims.", "palantir_relevance": "Authoritative SEC filing for Palantir business and risk context.", "quality_signal": "primary_source", "retrieval_tags": [ "10-k", "aip", "foundry", "investor", "ontology", "palantir", "revenue", "risk-factors", "sec" ], "triage_tier": "core", "triage_score": 90 }, { "id": "core-russell-norvig-ai-modern-approach", "title": "Artificial Intelligence: A Modern Approach", "authors_or_org": "Stuart Russell; Peter Norvig", "year": 2020, "source_type": "book", "bucket": "books", "url": "https://aima.cs.berkeley.edu", "doi_or_identifier": "isbn 9780134610993", "venue_or_site": "Pearson", "abstract_or_summary": "Standard AI textbook covering agents, search, logic, probabilistic reasoning, learning, natural language, and robotics.", "key_claims": [ "AI can be organized around rational agents that perceive, reason, learn, and act.", "Knowledge representation and reasoning are core components of intelligent systems." ], "ontology_relevance": "Places ontology/KR in the broader AI architecture of agents and action.", "ai_relevance": "Canonical AI reference for agent, logic, and learning perspectives.", "palantir_relevance": "", "quality_signal": "scholarly_book", "retrieval_tags": [ "agents", "ai", "book", "core", "knowledge-representation" ], "triage_tier": "core", "triage_score": 88 }, { "id": "pal-doc-aip-ethics-governance-2026", "title": "AI ethics and governance", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip/ethics-governance", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official guidance on AI ethics and governance in AIP, emphasizing interpretable tool delegation, debug visibility, testing, evaluation, and operational governance.", "key_claims": [ "AIP Logic tools can delegate tasks to interpretable components rather than relying only on LLM processing.", "Debug views can provide visibility into tool orchestration and handoffs.", "AIP Evals and modeling objectives support testing and governance.", "AIP Evals can evaluate AIP Logic functions across diverse test cases.", "Experiments can assess how input changes affect model responses.", "Data health monitoring is framed as a way to detect representativeness and quality issues." ], "ontology_relevance": "Frames ontology-linked tools as auditable components in AI systems.", "ai_relevance": "Relevant to enterprise AI governance, explainability, and production deployment.", "palantir_relevance": "Official governance narrative for AIP.", "quality_signal": "official_docs", "retrieval_tags": [ "ai-governance", "aip-evals", "aip-logic", "data-health", "debugging", "evals", "interpretable-tools", "palantir", "responsible-ai" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase3-pal-service-logs-debugging-2026", "title": "AIP observability: Service logs and debugging", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-observability/service-logs-and-debugging", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official page explaining log access and debugging permissions for traces and service logs in AIP observability.", "key_claims": [ "Administrators must enable log access for relevant projects to view traces and service logs.", "Users generally have access to logs for their own executions from the past 24 hours, with stack-specific caveats.", "Log access is permissioned rather than universally visible." ], "ontology_relevance": "Shows that observability itself is governed by permissions.", "ai_relevance": "Important for balancing accountability, privacy, and operational debugging in AI systems.", "palantir_relevance": "Primary source for log-access governance in AIP observability.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-observability", "debugging", "logs", "palantir", "permissions", "privacy" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase2-pal-aip-security-privacy-2026", "title": "AIP security and privacy", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip/aip-security", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation and FAQ describing Palantir's AIP security and privacy posture, including customer data protection, responsible processing, privacy, security, bias, and human judgment concerns.", "key_claims": [ "Palantir frames privacy and security as first principles for AIP deployment.", "The documentation acknowledges generative AI concerns around privacy, security, bias, discrimination, and human judgment.", "Security and privacy claims must be distinguished from independent audit evidence." ], "ontology_relevance": "Relevant to governance of ontology-backed operational AI systems that process sensitive data.", "ai_relevance": "Primary source for Palantir's stated AI security and privacy controls.", "palantir_relevance": "Important official governance source to contrast with public-sector critiques.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-security", "bias", "governance", "human-judgment", "palantir", "privacy" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase4-palantir-aip-virtual-tables-2024", "title": "AIP Virtual Tables", "authors_or_org": "Palantir Technologies", "year": 2024, "source_type": "blog", "bucket": "palantir", "url": "https://blog.palantir.com/aip-virtual-tables-5094b5e4b3bd", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir blog describing virtual tables as a way to expose external data to AIP and ontology workflows without always physically copying data.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "aip", "architecture", "data-integration", "palantir", "virtual-tables" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase4-palantir-logic-tools-rag-oag-2024", "title": "Building with Palantir AIP: Logic Tools for RAG/OAG", "authors_or_org": "Palantir Technologies", "year": 2024, "source_type": "blog", "bucket": "palantir", "url": "https://blog.palantir.com/building-with-palantir-aip-logic-tools-for-rag-oag-fdaf8938d02e", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir technical blog describing logic tools for retrieval-augmented and ontology-augmented generation, relevant to how AIP connects LLMs to governed operational logic.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "aip", "logic-tools", "oag", "ontology", "palantir", "rag" ], "triage_tier": "core", "triage_score": 86 }, { "id": "pal-doc-data-security-overview-2026", "title": "Data security overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/security/data-security-overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official Foundry security overview covering data access, permissions, and governance concepts that underpin ontology-backed workflows.", "key_claims": [ "Foundry security controls are designed to govern access to data and derived resources.", "Permissioning is a platform-level concern across data, applications, and ontology resources.", "Enterprise AI workflows depend on these controls when exposing data to tools or models." ], "ontology_relevance": "Places object security in the broader Foundry data-governance architecture.", "ai_relevance": "Relevant to claims that AIP can safely connect LLMs to sensitive enterprise data.", "palantir_relevance": "Official source for Foundry security framing.", "quality_signal": "official_docs", "retrieval_tags": [ "data-security", "foundry-security", "governance", "palantir", "permissions" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase5-palantir-foundry-platform-summary-llm-2026", "title": "Foundry platform summary for LLMs", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/getting-started/foundry-platform-summary-llm", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official Palantir documentation summarizing Foundry capabilities for LLMs, including Ontology MCP, AIP, tool use, permissions, and external agent integrations.", "key_claims": [ "Ontology MCP exposes application ontology resources as MCP tools for external AI agents.", "External agents can read objects, execute actions, and query data through configured ontology resources.", "Ontology MCP access is restricted through application restrictions and permissions and authenticates through OAuth 2.0 flows.", "The summary lists integrations with desktop agents such as Claude.ai, Microsoft Copilot Studio, Gemini Enterprise, and headless agent frameworks.", "MCP Hub provides a central location to discover and manage Ontology MCP servers configured across an enrollment." ], "ontology_relevance": "Compact official source for ontology as an external-agent interface, including resources, permissions, and tool discovery.", "ai_relevance": "Useful cross-reference for LLM architecture sections because it explicitly names agent platforms, SDKs, and authentication boundaries.", "palantir_relevance": "Official current Palantir platform summary for LLM-facing ontology capabilities.", "quality_signal": "official_docs", "retrieval_tags": [ "claude", "copilot-studio", "external-agents", "foundry", "gemini-enterprise", "llm", "mcp-hub", "oauth", "ontology-mcp", "palantir", "permissions", "phase5" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase4-palantir-functions-overview-2026", "title": "Functions: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official overview of Foundry Functions, which expose custom code and logic into applications, ontology workflows, and operational decision processes.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "architecture", "foundry", "functions", "logic", "ontology", "palantir" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase4-palantir-manage-published-functions-2026", "title": "Manage published functions", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/manage-functions", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for managing published functions, relevant to lifecycle and operational governance of reusable logic attached to ontology workflows.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "functions", "governance", "lifecycle", "palantir", "published-functions" ], "triage_tier": "core", "triage_score": 86 }, { "id": "pal-doc-object-security-overview-2026", "title": "Object security: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/security/object-security/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for object security in Foundry, describing how ontology object access can be governed through security controls and policies.", "key_claims": [ "Foundry supports security controls over ontology object access.", "Object security is a platform-level governance mechanism for application and workflow access.", "Security rules can shape what users and tools are allowed to see or use." ], "ontology_relevance": "Provides the governance layer that makes ontology primitives deployable in sensitive domains.", "ai_relevance": "Relevant because AIP agents and tools should inherit or respect object-level access boundaries.", "palantir_relevance": "Primary source on object-level security in Foundry.", "quality_signal": "official_docs", "retrieval_tags": [ "access-control", "foundry", "governance", "object-security", "palantir", "permissions" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase3-pal-ontology-manager-overview-2026", "title": "Ontology Manager: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-manager/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for Ontology Manager, the application used to build and maintain object types, action types, data connections, and ontology health.", "key_claims": [ "Ontology Manager supports creation and maintenance of object types and action types.", "It connects data to the Ontology and helps investigate updating behavior in user applications.", "Ontology management is treated as an ongoing operational process." ], "ontology_relevance": "Shows that ontology design is an administered lifecycle, not a one-time schema artifact.", "ai_relevance": "AI-facing objects and actions depend on the managed ontology resources exposed through this layer.", "palantir_relevance": "Documents the management application behind Foundry Ontology governance.", "quality_signal": "official_docs", "retrieval_tags": [ "action-types", "object-types", "ontology-lifecycle", "ontology-manager", "palantir" ], "triage_tier": "core", "triage_score": 86 }, { "id": "pal-doc-ontology-core-concepts-2026", "title": "Ontology: Core concepts", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/core-concepts", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Defines core ontology concepts: object types, objects, properties, link types, action types, and roles, with a comparison to datasets, rows, columns, and joins.", "key_claims": [ "An ontology maps datasets and models to object types, properties, link types, and action types.", "Object types represent real-world entities or events.", "Action types define changes or edits that users can apply.", "Properties correspond to object attributes, while link types represent relationships between object types.", "Action types define controlled changes or edits users can apply to objects, properties, and links.", "Object types represent real-world entities, events, or concepts.", "Link types encode relationships between object types.", "Action types define how object types can be modified." ], "ontology_relevance": "Gives the canonical public object/action model for Foundry Ontology.", "ai_relevance": "These primitives become structured context and controlled tools for AIP workflows.", "palantir_relevance": "Primary reference for Palantir's object/action schema.", "quality_signal": "official_docs", "retrieval_tags": [ "action-type", "link-type", "object-type", "ontology", "palantir", "property", "role", "roles" ], "triage_tier": "core", "triage_score": 86 }, { "id": "phase3-pal-mcp-security-2026", "title": "Palantir MCP: Security - Data governance", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/palantir-mcp/security", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official Palantir MCP security page describing data governance and data-flow/security models for MCP integration between Foundry and AI systems.", "key_claims": [ "Palantir MCP provides integration between AI systems and Foundry resources.", "Security and data-governance policies depend on how and where MCP is used.", "MCP use cases require attention to data-flow and security models." ], "ontology_relevance": "Frames ontology exposure as a governance and boundary-design issue.", "ai_relevance": "Important for external agent security, least privilege, and data-flow analysis.", "palantir_relevance": "Primary official source for MCP security framing.", "quality_signal": "official_docs", "retrieval_tags": [ "ai-systems", "data-governance", "foundry-resources", "mcp-security", "palantir" ], "triage_tier": "core", "triage_score": 86 }, { "id": "pal-blog-rag-oag-2023", "title": "Building with Palantir AIP: Data Tools for RAG / OAG", "authors_or_org": "Palantir Technologies", "year": 2023, "source_type": "technical_article", "bucket": "palantir", "url": "https://blog.palantir.com/building-with-palantir-aip-data-tools-for-rag-oag-b3b509c8b0f3", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Vendor-authored blog introducing Ontology Augmented Generation as a broader decision-centric extension of retrieval augmented generation using ontology data and relationships.", "key_claims": [ "OAG is presented as a decision-centric version of RAG.", "LLMs can retrieve context-specific business data such as orders, customers, and locations.", "Ontology context is framed as a way to reduce hallucination and improve relevance.", "Palantir distinguishes document retrieval from ontology-aware retrieval and tool use.", "Ontology Augmented Generation frames objects, relationships, and actions as AI context.", "The post is product narrative but useful for naming Palantir's OAG concept." ], "ontology_relevance": "Source for Palantir's OAG terminology and relationship to RAG.", "ai_relevance": "Connects ontology to generative AI grounding and retrieval patterns.", "palantir_relevance": "Palantir-authored explanation of OAG in AIP.", "quality_signal": "primary_source", "retrieval_tags": [ "aip", "llm", "llm-grounding", "oag", "ontology-augmented-generation", "palantir", "palantir-blog", "rag" ], "triage_tier": "core", "triage_score": 84 }, { "id": "pal-pcl-2026", "title": "Privacy and Civil Liberties", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "webpage", "bucket": "palantir", "url": "https://www.palantir.com/pcl", "doi_or_identifier": null, "venue_or_site": "Palantir", "abstract_or_summary": "Official Palantir page describing its privacy and civil liberties posture, including claims about preserving privacy and civil liberties while using data.", "key_claims": [ "Palantir positions privacy and civil liberties as part of its founding mission.", "The company claims its software supports controlled data use rather than selling customer data.", "The page is part of Palantir's response to surveillance and public trust concerns.", "Palantir presents privacy and civil liberties as part of its corporate operating model.", "The page emphasizes governance, legal compliance, and responsible use themes.", "Claims should be compared against public-sector contracts, parliamentary scrutiny, and civil-society critiques." ], "ontology_relevance": "Useful for understanding how governance and civil liberties are positioned around ontology-enabled data platforms.", "ai_relevance": "Provides official context for AI governance claims and trust posture.", "palantir_relevance": "Official counterpoint to external civil-liberties critiques.", "quality_signal": "primary_source", "retrieval_tags": [ "civil-liberties", "governance", "palantir", "privacy", "public-sector", "public-trust" ], "triage_tier": "core", "triage_score": 84 }, { "id": "pal-doc-action-types-2026", "title": "Action types", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/action-types", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for action types, the Ontology primitive used to define sanctioned changes to objects, properties, links, and related operational side effects.", "key_claims": [ "Action types define controlled modifications to ontology objects and links.", "Action definitions can include parameters, validation, and operational side effects.", "Action types are the primary public primitive for turning an ontology from a read model into an operational interface." ], "ontology_relevance": "Key source for Palantir's kinetic ontology model and governed write semantics.", "ai_relevance": "Action types are the natural tool boundary for AI agents that must write back to operations without arbitrary database access.", "palantir_relevance": "Official page on a central Foundry Ontology primitive.", "quality_signal": "official_docs", "retrieval_tags": [ "action-types", "governance", "ontology", "operational-actions", "palantir", "writeback" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-aip-analyst-2026", "title": "AIP Analyst: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/aip-analyst/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Documentation for AIP Analyst, an agentic interface that answers natural-language questions by searching ontology resources, creating object sets, transforming data, and generating outputs.", "key_claims": [ "AIP Analyst searches across the ontology for relevant data.", "It can create object sets and perform transformations before summarizing or visualizing results.", "It uses tools such as object type search, object type lookup, object search, and dataset lookup." ], "ontology_relevance": "Demonstrates how ontology metadata and object sets become navigable search space for AI analysis.", "ai_relevance": "Concrete example of agentic natural-language analysis over ontology data.", "palantir_relevance": "Official source for Palantir's ontology-grounded analyst agent.", "quality_signal": "official_docs", "retrieval_tags": [ "agentic-analysis", "aip-analyst", "object-sets", "ontology-search", "palantir" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase3-pal-aip-architecture-2026", "title": "AIP architecture overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/architecture-center/aip-architecture", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Architecture Center", "abstract_or_summary": "Official architecture page describing observability, logging for human and AI actions, ontology data flows, chained executions, and token/resource usage monitoring.", "key_claims": [ "AIP architecture includes monitoring for AI-driven workflows and agentic processes.", "The page describes logging for actions taken by human users or AI agents.", "Observability includes tracing cascades of chained executions and token/resource usage." ], "ontology_relevance": "Connects ontology data flows with platform architecture and governance controls.", "ai_relevance": "Strong source for operational AI controls beyond model prompting.", "palantir_relevance": "Official architecture-level statement about AI agent governance and observability.", "quality_signal": "official_docs", "retrieval_tags": [ "agentic-ai", "aip-architecture", "governance", "logging", "observability", "palantir" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-logic-compute-usage-2026", "title": "AIP Logic compute usage", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/logic/compute-usage", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for understanding and managing compute use in AIP Logic workflows.", "key_claims": [ "AIP Logic has measurable compute usage tied to workflow execution.", "Resource management is part of the production AI workflow story.", "Cost and capacity governance can be treated as operational controls, not only billing afterthoughts." ], "ontology_relevance": "Adds operational governance context to ontology-linked workflows.", "ai_relevance": "Important for managing LLM and workflow costs in production AI systems.", "palantir_relevance": "Official source for AIP Logic resource governance.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-logic", "compute-usage", "governance", "palantir", "resource-management" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase3-pal-aip-observability-trace-2026", "title": "AIP observability: Tracing", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-observability/trace-view", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official trace-view documentation explaining distributed traces across services, networks, process boundaries, and security boundaries.", "key_claims": [ "Trace view shows a visual timeline of workflow execution.", "Distributed traces can cross process, network, and security boundaries.", "Tracing helps understand the path a request takes through an application." ], "ontology_relevance": "Useful for tracing ontology-backed workflow execution across platform boundaries.", "ai_relevance": "Relevant to diagnosing chained agent/tool executions and accountability gaps.", "palantir_relevance": "Primary source for tracing mechanics in AIP observability.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-observability", "distributed-tracing", "palantir", "security-boundaries", "workflow-execution" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase5-palantir-mcp-hub-announcement-2026", "title": "Discover and manage Ontology MCP servers in MCP Hub", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/announcements/2026-05", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Announcements", "abstract_or_summary": "May 2026 Palantir announcement stating that Ontology MCP servers are discoverable through MCP Hub, with central management for MCP servers configured across an enrollment.", "key_claims": [ "Ontology MCP servers are discoverable through MCP Hub as of the 2026-05-21 announcement.", "Ontology MCP turns a Developer Console application into an MCP server for external AI agents.", "External AI agents can read object types, execute predefined action types, and run query functions, scoped to configured permissions.", "MCP Hub exposes server configuration details including tools and ontology resources.", "Palantir warns that enabling Ontology MCP makes ontology resources available to external MCP clients and should be checked against data-governance and security policies." ], "ontology_relevance": "Shows ontology resources becoming centrally discoverable and manageable as agent tool surfaces.", "ai_relevance": "Direct current evidence that external agent access is moving from individual tool configuration toward managed MCP server infrastructure.", "palantir_relevance": "Current official Palantir source for MCP Hub and Ontology MCP governance posture.", "quality_signal": "official_docs", "retrieval_tags": [ "data-governance", "external-agents", "mcp", "mcp-hub", "ontology-mcp", "palantir", "permissions", "phase5", "security", "tool-use" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-link-types-2026", "title": "Link types", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/link-types", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for defining link types between object types in the Ontology, including relationship semantics and navigation over connected operational objects.", "key_claims": [ "Link types define relationships between object types.", "Relationship modeling enables graph-style navigation across operational entities.", "Links make the Ontology more than isolated tables by encoding cross-object context." ], "ontology_relevance": "Primary source for the relationship layer in Palantir's operational ontology.", "ai_relevance": "Links supply graph context for agentic retrieval, object-set construction, and workflow reasoning.", "palantir_relevance": "Official documentation of a core Foundry Ontology primitive.", "quality_signal": "official_docs", "retrieval_tags": [ "link-types", "ontology", "operational-graph", "palantir", "relationships" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase2-pal-object-edit-history-2026", "title": "Object edits and materializations: Enable user edit history", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/object-edits/user-edit-history", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation on tracking history of user edits to objects indexed into Object Storage V2 and surfacing edit histories in Workshop modules or object views.", "key_claims": [ "Edit history is a concrete provenance mechanism for user-driven ontology object changes.", "Disabling edit history can permanently delete existing histories, making retention policy a governance issue.", "Operational ontology platforms need object-level change histories for auditability and trust." ], "ontology_relevance": "Shows provenance and audit needs for writable ontology objects.", "ai_relevance": "Relevant to AI-mediated edits where change history and accountability matter.", "palantir_relevance": "Primary source for Palantir object edit history behavior.", "quality_signal": "official_docs", "retrieval_tags": [ "audit", "edit-history", "object-edits", "object-storage", "palantir", "provenance" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase2-pal-object-edits-schema-migrations-2026", "title": "Object edits and materializations: Manage schema changes", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/object-edits/schema-migrations", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing breaking and non-breaking schema changes for object types, user edits, and Object Storage V2 schema migration support.", "key_claims": [ "Operational ontology schemas evolve, and breaking changes require migration support when user edits exist.", "Object Storage V2 provides a schema migration framework for certain ontology object changes.", "Schema governance is essential when ontology-backed workflows become writable and persistent." ], "ontology_relevance": "Concrete implementation material for ontology schema evolution and operational data quality.", "ai_relevance": "Important for AI systems whose actions depend on changing object schemas and stored edits.", "palantir_relevance": "Primary technical source for Palantir ontology schema migration behavior.", "quality_signal": "official_docs", "retrieval_tags": [ "object-edits", "object-storage", "ontology-governance", "palantir", "schema-migration", "validation" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-security-controls-2026", "title": "Object security: Security controls", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/security/object-security/security-controls", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing object security controls for constraining visibility or access to ontology objects and properties in Foundry.", "key_claims": [ "Object security controls can be configured to govern object visibility and access.", "Security controls are tied to ontology resources rather than only raw storage locations.", "Governance is part of the operational ontology model, not an external afterthought." ], "ontology_relevance": "Details the control layer surrounding ontology objects and properties.", "ai_relevance": "Security controls determine the data and actions available to AI applications and agents.", "palantir_relevance": "Official source for object security implementation concepts.", "quality_signal": "official_docs", "retrieval_tags": [ "governance", "object-security", "ontology-permissions", "palantir", "security-controls" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-object-types-2026", "title": "Object types", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/object-types", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for configuring object types in the Ontology, including properties, display metadata, backing data, and lifecycle considerations.", "key_claims": [ "Object types define classes of real-world entities, events, or concepts in Foundry Ontology.", "Object type configuration binds operational semantics to backing datasets or other data sources.", "Object type design shapes how applications, SDKs, and AI tools see enterprise data." ], "ontology_relevance": "Adds implementation detail to the object-type primitive beyond the high-level core concepts page.", "ai_relevance": "Object type definitions determine the typed data surface available to AIP Analyst, chatbots, SDKs, and MCP tools.", "palantir_relevance": "Primary official source on object type modeling in Foundry.", "quality_signal": "official_docs", "retrieval_tags": [ "data-modeling", "foundry", "object-types", "ontology", "palantir" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase3-pal-mcp-installation-2026", "title": "Palantir MCP: Installation", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/palantir-mcp/installation", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official installation page stating that Palantir MCP enables external AI systems to read Foundry data/metadata and interact via AI-friendly API endpoints using configured user-token permissions, while warning that external-system governance matters after data leaves Palantir.", "key_claims": [ "Palantir MCP uses configured user token permissions.", "External AI systems can read data and metadata from Foundry through AI-friendly APIs.", "Once data is accessed by an external system, governance of its use shifts toward that external system." ], "ontology_relevance": "Highlights boundary movement when ontology data and metadata are exposed outside Foundry.", "ai_relevance": "Essential source for risk analysis of MCP-mediated external agents.", "palantir_relevance": "Official admission of governance dependency on external systems after access.", "quality_signal": "official_docs", "retrieval_tags": [ "boundary", "data-governance", "external-agents", "mcp", "palantir", "permissions" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase3-pal-mcp-overview-2026", "title": "Palantir MCP: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/palantir-mcp/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official overview of Palantir MCP, which provides tools for external AI systems to take actions in Foundry, search or modify the ontology, and update Developer Console applications.", "key_claims": [ "Palantir MCP can search the ontology and safely modify the ontology.", "It can update Developer Console applications and generate OSDK-related changes.", "The toolset supports AI IDEs and external agents interacting with Foundry." ], "ontology_relevance": "Shows MCP reaching into ontology configuration, not only querying ontology data.", "ai_relevance": "Raises governance questions about external AI systems modifying platform configuration.", "palantir_relevance": "Primary official source for broader Palantir MCP capabilities.", "quality_signal": "official_docs", "retrieval_tags": [ "ai-ides", "foundry", "ontology-modification", "palantir", "palantir-mcp" ], "triage_tier": "core", "triage_score": 80 }, { "id": "pal-doc-platform-overview-2026", "title": "Platform overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/platform-overview/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Palantir platform overview describing AIP, Foundry, Apollo, and the ontology layer as an AI Mesh for connecting generative AI to operations.", "key_claims": [ "AIP connects generative AI to operations.", "Foundry and AIP include an ontology layer plus core services and security/governance layers.", "The Ontology is positioned as representing enterprise decisions, not only data.", "Foundry and AIP include an ontology layer plus core services and security and governance layers." ], "ontology_relevance": "Important source for platform-level ontology positioning.", "ai_relevance": "Connects ontology to the broader AIP/Foundry/Apollo architecture for operational AI.", "palantir_relevance": "Official high-level product architecture narrative.", "quality_signal": "official_docs", "retrieval_tags": [ "ai-mesh", "aip", "apollo", "foundry", "ontology-layer", "palantir", "platform-overview" ], "triage_tier": "core", "triage_score": 80 }, { "id": "phase5-palantir-connecting-agents-decisions-2026", "title": "Connecting Agents to Decisions", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "technical_article", "bucket": "palantir", "url": "https://blog.palantir.com/connecting-agents-to-decisions-277dee8ddb40", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir-authored technical narrative describing the Ontology as a decision-centric architecture for human-agent workflows, bringing data, logic, action, and security into a single operating model.", "key_claims": [ "Palantir frames the Ontology as a decision-centric architecture rather than a data-centric schema.", "The blog describes operational decisions as composed of data, logic, action, and security.", "The Ontology is presented as a way to surface enterprise logic assets as AI-ready tools for human-agent workflows.", "The blog claims ontology actions can be staged as scenarios, governed by granular controls, and written back to operational systems.", "Decision lineage and telemetry are presented as part of the memory and governance substrate for agents." ], "ontology_relevance": "High-value current source for the argument that operational ontology includes action, logic, security, lineage, and memory rather than only semantic labels.", "ai_relevance": "Connects Palantir's ontology architecture to agentic AI, tool use, operational memory, and governed writeback.", "palantir_relevance": "Current Palantir product narrative that should be labeled as vendor-authored but technically useful for architecture analysis.", "quality_signal": "primary_source", "retrieval_tags": [ "agent-memory", "agents", "decision-centric", "decision-lineage", "governed-action", "human-agent-workflows", "ontology", "palantir", "phase5", "security", "tool-use", "writeback" ], "triage_tier": "core", "triage_score": 78 }, { "id": "phase6-nhs-ndit-fdp-dpia-2026", "title": "FDP Data Protection Impact Assessment: NDIT Identifiable Version v3.0", "authors_or_org": "NHS England", "year": 2026, "source_type": "webpage", "bucket": "palantir", "url": "https://www.england.nhs.uk/wp-content/uploads/2025/08/redacted-ndit-nhs-england-fdp-dpia-identifiable-version-v3.0.pdf", "doi_or_identifier": "redacted-ndit-nhs-england-fdp-dpia-identifiable-version-v3.0", "venue_or_site": "NHS England", "abstract_or_summary": "Redacted DPIA for identifiable data processing in the FDP/NDIT context. It defines the FDP federation, Platform Contractor, NHS-PET, purposes, access controls, opt-out concepts, and the Ontology layer.", "key_claims": [ "The DPIA defines Palantir Technologies UK Ltd as Platform Contractor.", "It defines Ontology as mapping datasets and models to object types, properties, link types, and action types.", "It describes national and local FDP instances.", "It flags transparency actions around NDIT." ], "ontology_relevance": "High-value primary evidence because it defines ontology in operational NHS terms, not vendor marketing terms.", "ai_relevance": "Relevant to automated decision-making, AI enablement, and governance boundaries for model-linked workflows.", "palantir_relevance": "Names Palantir as platform contractor in a legally relevant privacy document.", "quality_signal": "primary_source", "retrieval_tags": [ "data-protection", "dpia", "fdp", "identifiable-data", "ndit", "nhs", "ontology-definition", "palantir", "phase6", "privacy" ], "triage_tier": "core", "triage_score": 78 }, { "id": "oa-https-openalex-org-w2108234081", "title": "Ontological foundations for structural conceptual models", "authors_or_org": "Giancarlo Guizzardi", "year": 2005, "source_type": "paper", "bucket": "academic", "url": "https://research.utwente.nl/en/publications/ontological-foundations-for-structural-conceptual-models", "doi_or_identifier": null, "venue_or_site": "University of Twente Research Information", "abstract_or_summary": "The main objective of this thesis is to contribute to the theory of Conceptual Modeling by proposing ontological foundations for structural conceptual models. Conceptual Modeling is a discipline of great importance to several areas in Computer Science. Its main objective is concerned with identifying, analyzing and describing the essential concepts and constraints of a universe of discourse, with the help of a (diagrammatic) modeling language that is based on a set of basic modeling concepts (forming a metamodel). In this thesis, we show how conceptual modeling languages can be evaluated and (re)designed with the purpose of improving their ontological adequacy. In simple terms, ontological adequacy is a measure of how close the models produced using a modeling language are to the situations in the reality they are supposed to represent. The thesis starts by proposing a systematic evaluation method for comparing a metamodel of the concepts underlying a language to a reference ontology of the corresponding domain in reality. The focus of this thesis is on general conceptual modeling languages (as opposed to domain specific ones). Hence, the proposed reference ontology is a foundational (or upper-level) ontology. Moreover, since, it focuses on structural modeling aspects (as opposed to dynamic ones), this foundational ontology is an ontology of objects, their properties and relations, their parts, the roles they play, and the types they instantiate. The proposed ontology was developed by adapting and extending a number of theories coming, primarily, from formal ontology in philosophy, but also from cognitive science and linguistics. Once developed, every subtheory of the ontology is used in the creation of methodological tools (e.g., modeling profiles, guidelines and design patterns). The expressiveness and relevance of these tools are shown throughout the thesis to solve some classical and recurrent conceptual modeling problems. Finally, the thesis demonstrates the applicability and usefulness of both the method and the proposed ontology by analyzing and extending a fragment of the Unified Modeling Language (UML) which deals with the construction of structural conceptual models.", "key_claims": [ "Conceptual models need ontological foundations to avoid category mistakes.", "OntoUML and related ideas help distinguish kinds, roles, phases, and relators." ], "ontology_relevance": "Key source for ontology-driven conceptual modeling quality.", "ai_relevance": "Useful for robust domain models and agent world models.", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "commercial", "conceptual-modeling", "core", "model-quality", "ontouml", "openalex" ], "triage_tier": "core", "triage_score": 77 }, { "id": "phase15-w3c-direct-mapping-relational-data-rdf-2012", "title": "A Direct Mapping of Relational Data to RDF", "authors_or_org": "Marcelo Arenas; Alexandre Bertails; Eric Prud'hommeaux; Juan Sequeda; W3C RDB2RDF Working Group", "year": 2012, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/rdb-direct-mapping", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Direct Mapping is the W3C Recommendation for defining a default RDF graph representation of relational database content and schema. It complements R2RML by providing a canonical baseline for relational-to-RDF translation.", "key_claims": [ "Direct Mapping defines a default RDF representation for relational databases.", "The recommendation makes table, row, primary-key, foreign-key, and literal translation rules explicit.", "Default mappings provide a baseline before customized R2RML-style semantic mappings are designed." ], "ontology_relevance": "Adds the baseline relational-to-RDF mapping layer needed to explain how existing enterprise databases can become semantic graph material.", "ai_relevance": "Direct mappings matter for AI data infrastructure because they expose how relational structures can become graph-addressable without ad hoc prompt-level interpretation.", "palantir_relevance": "Useful background for Palantir-style ontology integration: it separates physical database structure from semantic object modeling.", "quality_signal": "official_standard", "retrieval_tags": [ "data-integration", "direct-mapping", "knowledge-graph-construction", "mapping-language", "phase15", "rdf", "relational-data", "w3c" ], "triage_tier": "core", "triage_score": 76 }, { "id": "phase15-dataspace-protocol-2025-1", "title": "Dataspace Protocol 2025-1", "authors_or_org": "Eclipse Dataspace Protocol project; International Data Spaces Association contributors", "year": 2025, "source_type": "specification", "bucket": "technical", "url": "https://eclipse-dataspace-protocol-base.github.io/DataspaceProtocol/2025-1", "doi_or_identifier": null, "venue_or_site": "Eclipse Dataspace Protocol Specification", "abstract_or_summary": "The Dataspace Protocol 2025-1 specification defines schemas and protocols for entities to publish data, negotiate agreements, and access data as part of federated dataspaces. It builds on Web technologies and usage-control-oriented interoperability.", "key_claims": [ "The Dataspace Protocol defines interoperable schemas and protocols for federated data sharing.", "It covers publication, agreement negotiation, and data access in dataspace federations.", "Usage control and Web technologies are central to cross-organization data interoperability." ], "ontology_relevance": "Connects semantic catalogs, usage policies, and protocol-level interoperability for ontology-backed AI systems that cross organizational boundaries.", "ai_relevance": "Provides current protocol evidence for agentic AI and enterprise systems that must discover data catalogs, negotiate usage, and access resources across organizational boundaries.", "palantir_relevance": "Relevant comparator for Palantir Ontology MCP and public-sector data platforms because it frames data access as federated agreement and protocol, not only platform integration.", "quality_signal": "official_standard", "retrieval_tags": [ "agreement-negotiation", "data-space", "dataspace-protocol", "dataspaces", "federated-data-sharing", "phase15", "usage-control", "usage-policy" ], "triage_tier": "core", "triage_score": 76 }, { "id": "phase16-oasis-xacml-3-policy-access-control-2013", "title": "eXtensible Access Control Markup Language (XACML) Version 3.0", "authors_or_org": "OASIS eXtensible Access Control Markup Language Technical Committee", "year": 2013, "source_type": "standard", "bucket": "technical", "url": "https://docs.oasis-open.org/xacml/3.0/xacml-3.0-core-spec-os-en.html", "doi_or_identifier": "oasis standard, 22 january 2013", "venue_or_site": "OASIS Standard", "abstract_or_summary": "XACML 3.0 defines a policy language and request/response model for attribute-based access control, including policies, rules, targets, obligations, advice, combining algorithms, and decision points.", "key_claims": [ "Access-control decisions can be represented as machine-readable policies and request/response evaluations.", "Policies can include obligations and advice that accompany decisions.", "Agentic AI governance needs policy enforcement layers outside the LLM when tools can read or change state." ], "ontology_relevance": "Strengthens the policy-governance lane with a mature access-control specification adjacent to semantic usage policies.", "ai_relevance": "Tool-using agents and ontology-exposed actions require policy enforcement beyond model prompts; XACML is a mature policy-control comparator alongside ODRL, SHACL, and deontic policy work.", "palantir_relevance": "Comparator for Palantir action permissions, dynamic security, and external-agent tool governance.", "quality_signal": "official_standard", "retrieval_tags": [ "access-control", "attribute-based-access-control", "governed-action", "oasis", "ontology-governance", "phase16", "policy-enforcement", "runtime-governance", "xacml" ], "triage_tier": "core", "triage_score": 76 }, { "id": "phase16-iec-common-data-dictionary-cdd", "title": "IEC Common Data Dictionary (IEC CDD)", "authors_or_org": "International Electrotechnical Commission", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://cdd.iec.ch", "doi_or_identifier": "iec 61360 / iec common data dictionary", "venue_or_site": "International Electrotechnical Commission", "abstract_or_summary": "IEC CDD is the IEC Common Data Dictionary for electrotechnical product classes and properties. It provides standardized machine-readable product semantics for components, properties, and classification used across industrial and standards ecosystems.", "key_claims": [ "A common data dictionary standardizes product classes, properties, and definitions for electrotechnical domains.", "Industrial interoperability requires shared identifiers and property semantics across tools and organizations.", "Ontology-backed AI agents can use data dictionaries to ground asset and component terms." ], "ontology_relevance": "Adds industrial data dictionary governance and standardized product/property semantics to the KB.", "ai_relevance": "Industrial AI systems that reason about assets, components, and product properties need standardized dictionaries to avoid ambiguous labels and incompatible schemas.", "palantir_relevance": "Useful comparator for operational asset ontologies in industrial enterprise settings.", "quality_signal": "official_standard", "retrieval_tags": [ "common-data-dictionary", "data-dictionary", "iec", "industrial-semantic-standards", "ontology-governance", "phase16", "product-classification", "semantic-interoperability" ], "triage_tier": "core", "triage_score": 76 }, { "id": "phase4-std-w3c-r2rml-2012", "title": "R2RML: RDB to RDF Mapping Language", "authors_or_org": "Souripriya Das; Seema Sundara; Richard Cyganiak; W3C RDB2RDF Working Group", "year": 2012, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/r2rml", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "R2RML is the W3C Recommendation for expressing customized mappings from relational databases to RDF datasets. It defines logical tables, subject maps, predicate-object maps, joins, and graph maps for generating RDF from relational source data.", "key_claims": [ "R2RML standardizes relational-to-RDF mappings for customized graph generation.", "Mapping rules make the connection between operational tables and semantic graph entities explicit.", "R2RML is a key standards bridge between existing databases and ontology-backed knowledge graphs." ], "ontology_relevance": "Supports the claim that ontology-backed AI should preserve explicit mappings from physical data schemas to semantic objects and relations.", "ai_relevance": "AI systems that rely on ontology/KG layers often need traceable mappings from existing operational databases; R2RML is a standards anchor for that mapping layer.", "palantir_relevance": "A neutral comparator for enterprise ontology systems that map operational database tables into object/relationship layers.", "quality_signal": "official_standard", "retrieval_tags": [ "data-integration", "knowledge-graph", "knowledge-graph-construction", "mapping-language", "mappings", "phase15", "r2rml", "rdf", "relational-data", "w3c" ], "triage_tier": "core", "triage_score": 76 }, { "id": "p2-llm-ont-033", "title": "RDF 1.2 Concepts and Abstract Data Model", "authors_or_org": "World Wide Web Consortium", "year": 2026, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/rdf12-concepts", "doi_or_identifier": "w3c cr-rdf12-concepts-20260407", "venue_or_site": "W3C Candidate Recommendation Snapshot", "abstract_or_summary": "Current W3C Candidate Recommendation snapshot for RDF 1.2 concepts and abstract data model, updating the RDF graph model used by semantic web systems.", "key_claims": [ "RDF remains the foundational graph data model for interoperable semantic data.", "Standard abstract graph semantics matter for exchange across tools and organizations.", "RDF 1.2 work shows active maintenance of the Semantic Web standards stack." ], "ontology_relevance": "Base graph model for OWL, SKOS, SHACL, DCAT, and many ontology-backed knowledge graphs.", "ai_relevance": "Provides a standards-based substrate for graph retrieval, provenance, and machine-readable context.", "palantir_relevance": "Reference point for comparing proprietary operational ontology graphs with open semantic graph standards.", "quality_signal": "official_standard", "retrieval_tags": [ "graph-data", "rdf-1-2", "semantic-web", "standard", "w3c" ], "triage_tier": "core", "triage_score": 76 }, { "id": "phase16-w3c-sparql11-federated-query-2013", "title": "SPARQL 1.1 Federated Query", "authors_or_org": "W3C SPARQL Working Group", "year": 2013, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/sparql11-federated-query", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "SPARQL 1.1 Federated Query extends SPARQL with SERVICE-based subqueries over remote endpoints. It provides a standards route for querying distributed RDF data without materializing every graph into one store.", "key_claims": [ "SPARQL federated query allows portions of a query to be delegated to remote SPARQL services.", "Federated query supports distributed semantic architectures where data remains under separate endpoints.", "Cross-organization ontology-backed AI can use federated query patterns as one retrieval mechanism." ], "ontology_relevance": "Adds the federated query layer to the semantic-web standards stack and data-space discussion.", "ai_relevance": "Federated retrieval and agentic question answering over organizational data require controlled distributed query mechanisms, provenance, and endpoint boundaries rather than uncontrolled data copying.", "palantir_relevance": "Useful neutral comparator for federated semantic access; not evidence about Palantir architecture.", "quality_signal": "official_standard", "retrieval_tags": [ "distributed-query", "federated-data-sharing", "federated-query", "phase16", "rdf", "semantic-interoperability", "sparql", "w3c" ], "triage_tier": "core", "triage_score": 76 }, { "id": "pal-doc-chatbot-studio-2026", "title": "AIP Chatbot Studio: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/chatbot-studio/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Documentation for building AIP Chatbots, formerly AIP Agents, powered by LLMs, the Ontology, documents, and custom tools, with internal and external deployment paths.", "key_claims": [ "AIP Chatbots are equipped with enterprise-specific information and tools.", "They are powered by LLMs, the Ontology, documents, and custom tools.", "They can support dynamic, context-aware read and write workflows." ], "ontology_relevance": "Shows ontology as a source of structured context and tools for conversational agents.", "ai_relevance": "Primary source for Palantir's LLM agent builder model.", "palantir_relevance": "Official documentation for AIP Chatbot Studio/Agent Studio evolution.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-agent-studio", "aip-chatbot-studio", "llm-agents", "palantir", "read-write-workflows" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase3-pal-aip-evals-overview-2026", "title": "AIP Evals: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-evals/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official overview of AIP Evals as a testing environment for AIP Logic, chatbot, and code-authored functions, designed for nondeterministic LLM behavior.", "key_claims": [ "AIP Evals supports test cases and evaluation functions.", "It compares results against previous versions of a function.", "It is designed to address nondeterminism in LLM applications." ], "ontology_relevance": "Provides a control layer for ontology-facing workflows that may change over time.", "ai_relevance": "Key evidence for model/workflow evaluation rather than one-off prompt testing.", "palantir_relevance": "Primary source for AIP evaluation mechanics.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-evals", "llm-evaluation", "palantir", "regression", "test-cases" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-aip-features-2026", "title": "AIP features", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/aip/aip-features", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation listing AIP application and builder capabilities, including LLM-backed workflows, agents, applications, OSDK, AIP Logic, and Palantir MCP.", "key_claims": [ "AIP enables developers to build LLM-backed workflows, agents, and applications.", "Developer tooling provides access to ontology data, logic, and actions.", "Palantir MCP connects external AI IDEs and agents to ontology and Foundry context." ], "ontology_relevance": "Links ontology resources to developer and agent toolchains.", "ai_relevance": "Shows how AIP exposes ontology data/actions to LLM-backed software.", "palantir_relevance": "Official feature map for AIP and related developer capabilities.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "aip-features", "llm-workflows", "ontology-data", "palantir", "palantir-mcp" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-logic-metrics-2026", "title": "AIP Logic metrics", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/logic/aip-logic-metrics", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for metrics associated with AIP Logic, supporting monitoring of workflow behavior and resource use.", "key_claims": [ "AIP Logic exposes metrics that can be used to monitor logic function execution.", "Metrics make AI workflow behavior observable beyond prompt outputs.", "Execution metrics support governance, performance analysis, and operational troubleshooting." ], "ontology_relevance": "Supports the claim that ontology-linked logic has observable operational behavior.", "ai_relevance": "Relevant to AI observability, evaluation, and operational monitoring.", "palantir_relevance": "Official evidence for runtime metrics in AIP Logic.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-logic", "metrics", "observability", "palantir", "workflow-monitoring" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-logic-core-concepts-2026", "title": "AIP Logic: Core concepts", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/logic/core-concepts", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official concept page for AIP Logic, covering logic functions, blocks, inputs, outputs, tools, and execution concepts.", "key_claims": [ "AIP Logic workflows are composed from blocks with typed inputs and outputs.", "LLM calls can be combined with tools and deterministic business logic.", "Logic concepts create a bridge between AI generation and governed operational execution." ], "ontology_relevance": "Connects ontology entities/actions to reusable workflow logic.", "ai_relevance": "Provides details for how Palantir packages AI workflow composition rather than only chat interfaces.", "palantir_relevance": "Official source for AIP Logic primitives.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-logic", "blocks", "palantir", "tools", "typed-inputs", "workflow-composition" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-logic-overview-2026", "title": "AIP Logic: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/logic/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official overview of AIP Logic, Palantir's tool for creating logic functions and AI workflows using blocks, models, tools, and ontology-connected inputs and outputs.", "key_claims": [ "AIP Logic is a builder surface for AI-powered workflows and logic functions.", "Logic workflows can combine LLM calls, tools, ontology data, and deterministic processing.", "Logic functions can be reused by applications, agents, and other Foundry components.", "AIP Logic functions can be automated.", "Ontology edits can be automatically applied or staged for human review.", "AIP Logic is positioned as a workflow layer for AI-assisted operations." ], "ontology_relevance": "Shows how executable logic is attached to ontology-backed operations.", "ai_relevance": "Primary source for Palantir's mechanism for composing LLM workflows with controlled tools and data.", "palantir_relevance": "Official source for AIP Logic as a core AIP mechanism.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-logic", "human-review", "llm-tools", "logic-functions", "ontology", "ontology-edits", "palantir", "workflow", "workflows" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase4-palantir-model-catalog-overview-2026", "title": "AIP Model Catalog: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/model-catalog/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for Model Catalog, the Foundry service used to register and manage AI models available to AIP workflows.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "aip", "architecture", "model-catalog", "model-governance", "palantir" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase3-pal-aip-observability-run-history-2026", "title": "AIP observability: Execution history", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-observability/run-history", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for run history of functions, actions, and automations, including recent executions in a resource-level run-history tab.", "key_claims": [ "Run history is available for a Function, Action, or automation.", "The run-history tab shows recent executions.", "Execution history gives a resource-level view of operational workflow runs." ], "ontology_relevance": "Applies audit-like traceability to actions and functions attached to the Ontology.", "ai_relevance": "Important for reviewing what AI-assisted workflows actually executed.", "palantir_relevance": "Primary evidence for execution-history controls.", "quality_signal": "official_docs", "retrieval_tags": [ "actions", "auditability", "automations", "functions", "palantir", "run-history" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-functions-2026", "title": "Functions in the Ontology", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/functions", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing functions as logic attached to the Ontology and usable by applications, object views, SDKs, and workflows.", "key_claims": [ "Functions define reusable logic over ontology resources.", "Functions can extend ontology behavior beyond stored properties and links.", "Functions provide a controlled logic surface for applications and AI-enabled workflows." ], "ontology_relevance": "Shows how Palantir attaches computation and business logic to ontology entities.", "ai_relevance": "Functions are a tool surface for AIP Logic, agents, SDKs, and evaluations.", "palantir_relevance": "Official source for ontology-linked logic.", "quality_signal": "official_docs", "retrieval_tags": [ "business-logic", "functions", "ontology", "palantir", "tools" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase3-pal-functions-on-objects-2026", "title": "Functions on objects (FOO)", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/functions-on-objects", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing functions that operate natively on ontology objects and object sets, including retrieval, storage, modification, lineage, and transparency claims.", "key_claims": [ "Functions can take object and object-set types as parameters.", "Functions can search object sets and modify objects using OntologyEditFunctions.", "Palantir states functions go beyond FaaS by adding native ontology support, data security, lineage, and transparency." ], "ontology_relevance": "Shows how business logic attaches directly to ontology objects and object sets.", "ai_relevance": "AIP workflows can combine LLM steps with deterministic functions over typed operational objects.", "palantir_relevance": "Strong technical bridge between Ontology and executable logic.", "quality_signal": "official_docs", "retrieval_tags": [ "functions", "lineage", "object-sets", "ontology-objects", "palantir", "transparency" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase3-pal-project-permissions-ontology-2026", "title": "Migrate to project-based permissions", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-manager/migrate-to-project-based-permissions", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official page explaining project-based permissions for ontology resources such as object types, action types, link types, interfaces, and shared properties.", "key_claims": [ "Ontology resources can be saved within projects and inherit project permissions.", "Object and link instance permissions remain dependent on backing datasource location.", "Permissions to view, edit, and manage ontology resources are administered through Compass." ], "ontology_relevance": "Clarifies governance boundaries between ontology metadata and instance-level data access.", "ai_relevance": "Relevant to which ontology resources AI agents and apps can discover, edit, or use.", "palantir_relevance": "Primary source for ontology resource permissioning.", "quality_signal": "official_docs", "retrieval_tags": [ "compass", "governance", "ontology-permissions", "palantir", "project-permissions" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase2-pal-ontology-mcp-auth-2026", "title": "Ontology MCP: Authentication and authorization", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-mcp/authentication-and-authorization", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation explaining how Ontology MCP uses the OAuth 2.0 configuration, application restrictions, permissions, and scoped tokens of the Developer Console application.", "key_claims": [ "Ontology MCP uses existing OAuth 2.0 application configuration rather than a separate authentication system.", "Application restrictions and permissions apply to MCP requests made by external agents.", "Scoped tokens define which ontology operations and resources an MCP client can access." ], "ontology_relevance": "Shows how ontology access can be governed when exposed as agent tools.", "ai_relevance": "Directly relevant to secure tool use and agent authorization.", "palantir_relevance": "Important primary source for Palantir's safety boundary around external AI agents.", "quality_signal": "official_docs", "retrieval_tags": [ "governed-action", "mcp", "oauth", "ontology-mcp", "palantir", "permissions", "security" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase5-palantir-ontology-mcp-getting-started-2026", "title": "Ontology MCP: Getting started", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology-mcp/getting-started", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official getting-started guide for Ontology MCP, including discovery through MCP Hub and configuration of MCP servers for ontology resources.", "key_claims": [ "Ontology MCP servers are discoverable through the MCP Hub application.", "MCP Hub lists MCP servers configured on an enrollment and links to their Developer Console applications.", "Ontology MCP server details include the tools and ontology resources exposed to agents.", "The guide positions MCP Hub as a workflow for managing external agent access to ontology resources." ], "ontology_relevance": "Adds implementation-level detail for discovering and managing ontology-backed MCP servers.", "ai_relevance": "Useful for tracing how ontology resources become operational agent tools and how admins inspect those tool surfaces.", "palantir_relevance": "Official Palantir documentation for the current Ontology MCP workflow.", "quality_signal": "official_docs", "retrieval_tags": [ "agent-configuration", "developer-console", "external-agents", "mcp-hub", "ontology-mcp", "palantir", "phase5", "tool-use" ], "triage_tier": "core", "triage_score": 74 }, { "id": "phase3-pal-ontology-mcp-tools-config-2026", "title": "Ontology MCP: MCP tools and agent configuration", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-mcp/mcp-tools-and-agent-configuration", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation explaining how ontology actions can be described for AI agents through tool descriptions and configuration fields.", "key_claims": [ "Ontology Manager includes an Agent tool description field for actions exposed to agents.", "Tool descriptions guide AI agents on when and how to use each action.", "Action configuration becomes part of the agent-facing contract." ], "ontology_relevance": "Shows that ontology action metadata is adapted for agent tool semantics.", "ai_relevance": "Important for prompt/tool-instruction governance and misuse prevention.", "palantir_relevance": "Primary source for configuring agent-facing action tools.", "quality_signal": "official_docs", "retrieval_tags": [ "actions", "agent-configuration", "ontology-mcp", "palantir", "tool-description" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-ontology-mcp-2026", "title": "Ontology MCP: Sample architecture", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology-mcp/sample-architecture", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Documentation describing Ontology MCP as a way to expose object types, action types, and query functions from a Developer Console application as tools for agents.", "key_claims": [ "Ontology MCP exposes object types, action types, and query functions to agents.", "Object types can be reachable through SQL tools.", "Action types can become individual MCP tools for controlled writes." ], "ontology_relevance": "Shows ontology resources being translated into machine-callable tools.", "ai_relevance": "Directly relevant to external agents and MCP-based AI tool use.", "palantir_relevance": "Primary source for Palantir's agent-tool interface over ontology primitives.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "controlled-actions", "mcp", "ontology-mcp", "palantir", "tool-use" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-osdk-typescript-2026", "title": "Ontology SDK: TypeScript OSDK", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-sdk/typescript-osdk", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Documentation for generated TypeScript SDKs that expose ontology object types, actions, and functions to application developers.", "key_claims": [ "Action types in the Ontology become predefined operations in generated SDK code.", "The OSDK exposes objects, actions, and functions for application development.", "Functions can define custom logic on ontology data.", "Generated SDK code exposes object types, action types, and functions from the Ontology.", "Action types become predefined operations in application code.", "OSDK is a developer surface for ontology-backed applications.", "TypeScript OSDK documentation is generated around ontology objects, actions, and queries.", "Developer Console can generate documentation based on a specific Ontology.", "The OSDK gives application developers typed access to operational resources." ], "ontology_relevance": "Shows how ontology primitives become programming abstractions.", "ai_relevance": "OSDK is one integration path for AIP-powered applications and external AI-enabled tools.", "palantir_relevance": "Concrete developer-tool evidence for ontology-backed app building.", "quality_signal": "official_docs", "retrieval_tags": [ "actions", "functions", "ontology", "ontology-sdk", "osdk", "palantir", "queries", "typed-client", "typescript", "typescript-osdk" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-action-permissions-2026", "title": "Permission checks when applying an Action", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/object-edits/permission-checks", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation describing permission checks for Actions, including different behavior for single-datasource and multi-datasource object types and submission criteria.", "key_claims": [ "Action permission checks depend on object backing architecture.", "For single-datasource edits, users need object view/load access and must pass action submission criteria.", "Creating new objects requires access to the input datasource." ], "ontology_relevance": "Connects ontology actions to access control and governance.", "ai_relevance": "Provides evidence that AI-triggered or user-triggered actions can be bounded by platform permissions and action criteria.", "palantir_relevance": "Concrete source for action governance and authorization semantics.", "quality_signal": "official_docs", "retrieval_tags": [ "access-control", "action-permissions", "ontology-security", "palantir", "submission-criteria" ], "triage_tier": "core", "triage_score": 74 }, { "id": "pal-doc-why-ontology-2026", "title": "Why create an Ontology?", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology/why-ontology", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Explains the value proposition for an ontology in Foundry: decision-centric modeling with data, logic, action, and security as the elements of operational decisions.", "key_claims": [ "Palantir models operational decisions as data, logic, action, and security.", "The decision-centric Ontology connects humans and agents to operations.", "The Ontology is framed as more than data cataloging or schema design.", "The Ontology connects humans and agents to operational decisions.", "The Ontology is framed as more than cataloging data or designing schemas.", "Palantir frames operational decisions around data, logic, action, and security.", "The Ontology is positioned as more than data cataloging or schema design." ], "ontology_relevance": "Important source for the decision-centric interpretation of Palantir's ontology practice.", "ai_relevance": "Directly connects ontology to human-agent operational workflows.", "palantir_relevance": "Official framing of why the ontology matters commercially and operationally.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "data-logic-action-security", "decision-centric", "operations", "palantir" ], "triage_tier": "core", "triage_score": 74 }, { "id": "industry-commercial-allegrograph-llm", "title": "AllegroGraph knowledge graph, vector, and LLM materials", "authors_or_org": "Franz Inc.", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://allegrograph.com", "doi_or_identifier": null, "venue_or_site": "AllegroGraph / Franz Inc.", "abstract_or_summary": "AllegroGraph materials position RDF knowledge graphs, reasoning, vector storage, and LLM/RAG capabilities as a combined neuro-symbolic AI platform.", "key_claims": [ "RDF graph storage can be combined with vector similarity and LLM workflows.", "Reasoning and semantic relationships can provide explainable structure for AI context.", "Neuro-symbolic positioning joins curated symbols with statistical retrieval." ], "ontology_relevance": "Shows a standards-based graph database extending ontology stores toward vector and LLM workloads.", "ai_relevance": "Relevant to hybrid symbolic-vector retrieval and GraphRAG architectures.", "palantir_relevance": "Useful comparator for AI over graph semantics outside Palantir.", "quality_signal": "official_docs", "retrieval_tags": [ "allegrograph", "graphrag", "llm", "neuro-symbolic", "rdf", "vector-search" ], "triage_tier": "core", "triage_score": 72 }, { "id": "pal-blog-connecting-ai-decisions-2024", "title": "Connecting AI to Decisions with the Palantir Ontology", "authors_or_org": "Akshay Krishnaswamy, Palantir", "year": 2024, "source_type": "technical_article", "bucket": "palantir", "url": "https://blog.palantir.com/connecting-ai-to-decisions-with-the-palantir-ontology-c73f7b0a1a72", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Vendor-authored blog explaining Palantir's decision-centric ontology thesis and why Palantir argues operational AI depends on modeling decisions, actions, and context rather than only data.", "key_claims": [ "The Ontology is designed to represent decisions in an enterprise, not simply data.", "Traditional data architectures do not capture reasoning or action well enough for operational AI.", "AIP's operational impact is attributed to software architecture around the Ontology.", "Palantir argues that AI must be connected to operational decisions rather than isolated chat interfaces.", "The Ontology is presented as the structure that maps enterprise data, logic, action, and security to decision workflows.", "The post is useful conceptual framing, but vendor-authored." ], "ontology_relevance": "Strong conceptual source for Palantir's ontology practice and decision-centric framing.", "ai_relevance": "Explains Palantir's argument that ontology is the bridge from AI to operational decisions.", "palantir_relevance": "High-value vendor narrative by a Palantir Chief Architect.", "quality_signal": "primary_source", "retrieval_tags": [ "aip", "blog", "decision-centric-ontology", "decisions", "enterprise-decisions", "ontology", "operational-ai", "palantir", "palantir-blog" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase4-std-dpv-2-0-2024", "title": "Data Privacy Vocabulary (DPV) Version 2.0", "authors_or_org": "W3C Data Privacy Vocabularies and Controls Community Group", "year": 2024, "source_type": "community_standard", "bucket": "technical", "url": "https://www.w3.org/community/reports/dpvcg/CG-FINAL-dpv-20240801", "doi_or_identifier": null, "venue_or_site": "W3C Community Final Specification", "abstract_or_summary": "The Data Privacy Vocabulary 2.0 provides a machine-readable vocabulary for personal data handling, purposes, processing operations, legal bases, rights, risks, controls, and organizational measures.", "key_claims": [ "Privacy and data processing metadata can be expressed with shared machine-readable vocabularies.", "DPV gives taxonomies for purposes, processing operations, personal data categories, legal bases, rights, risks, and controls.", "AI context and retrieval layers can use privacy vocabularies to constrain use and disclosure." ], "ontology_relevance": "Important vocabulary for privacy-aware ontology governance and compliance metadata.", "ai_relevance": "Ontology-backed AI needs explicit privacy context: lawful basis, purpose limits, sensitive data categories, processing operations, rights, and controls should be retrievable and checkable by agents.", "palantir_relevance": "Neutral comparator for health-data and public-sector governance analysis; not Palantir-specific validation.", "quality_signal": "official_docs", "retrieval_tags": [ "compliance", "data-governance", "dpv", "gdpr", "ontology-governance", "phase16", "policy-vocabulary", "privacy", "privacy-vocabulary", "risk", "semantic-interoperability", "w3c-community" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-eccenca-corporate-memory", "title": "eccenca Corporate Memory knowledge graph and data product governance materials", "authors_or_org": "eccenca", "year": 2026, "source_type": "docs", "bucket": "commercial", "url": "https://eccenca.com/products/corporate-memory", "doi_or_identifier": null, "venue_or_site": "eccenca", "abstract_or_summary": "eccenca Corporate Memory materials cover knowledge graph lifecycle management, semantic data governance, SHACL validation, metadata, and AI-ready semantic context.", "key_claims": [ "Knowledge graphs require lifecycle workflows for modeling, validating, publishing, and reuse.", "Semantic data products need provenance, stewardship, and quality controls.", "SHACL and governance workflows can make ontology rules enforceable." ], "ontology_relevance": "Useful source for operationalizing ontology lifecycle management and semantic governance.", "ai_relevance": "AI systems can use governed knowledge graph assets as trusted contextual memory.", "palantir_relevance": "Non-Palantir example of corporate memory and ontology governance for operations and AI.", "quality_signal": "official_docs", "retrieval_tags": [ "ai-ready-kg", "corporate-memory", "data-products", "eccenca", "semantic-governance", "shacl" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase16-eclass-rdf-owl-product-classification", "title": "eCl@ss in RDF/OWL", "authors_or_org": "eCl@ss e.V.", "year": 2025, "source_type": "docs", "bucket": "technical", "url": "https://www.eclass.eu/en/standard/eclassrdfowl", "doi_or_identifier": "ecl@ss rdf/owl distribution", "venue_or_site": "eCl@ss Standard Documentation", "abstract_or_summary": "The eCl@ss RDF/OWL representation exposes the eCl@ss product and service classification system as semantic-web resources. It supports product classification, properties, identifiers, and machine-readable product semantics across procurement and industry contexts.", "key_claims": [ "Product and service classifications can be represented in RDF/OWL for machine-readable interoperability.", "Standard product semantics support procurement, catalog integration, and supply-chain data exchange.", "Ontology-backed AI in operations benefits from shared product classes and property definitions." ], "ontology_relevance": "Adds industrial/product classification infrastructure to the ontology and enterprise AI evidence base.", "ai_relevance": "Enterprise agents working with products, supply chains, procurement, or asset catalogs need standardized product semantics instead of ad hoc item descriptions.", "palantir_relevance": "Comparator for supply-chain and asset ontology modeling; not Palantir-specific evidence.", "quality_signal": "official_docs", "retrieval_tags": [ "eclass", "industrial-semantic-standards", "ontology-governance", "phase16", "product-classification", "rdf", "semantic-interoperability", "supply-chain" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-ontotext-graphdb-docs", "title": "GraphDB documentation: reasoning, SHACL, and semantic graph database capabilities", "authors_or_org": "Ontotext", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://graphdb.ontotext.com/documentation", "doi_or_identifier": null, "venue_or_site": "Ontotext GraphDB Documentation", "abstract_or_summary": "Official documentation for GraphDB, an RDF database with SPARQL, inferencing, SHACL validation, repository management, and semantic retrieval features.", "key_claims": [ "RDF graph stores can apply inferencing to derive implicit relationships and classifications.", "SHACL validation can enforce graph quality and business constraints.", "Semantic retrieval can combine graph structure with text and similarity features.", "RDF graph stores can apply inference to derive implicit classifications and relationships.", "Semantic retrieval can combine graph structure, text, similarity, and provenance." ], "ontology_relevance": "Strong source for ontology execution through RDF, SPARQL, OWL/RDFS-style reasoning, and SHACL validation.", "ai_relevance": "Useful for GraphRAG and semantic search patterns that need validated, ontology-aware context.", "palantir_relevance": "Provides a standards-based alternative to proprietary ontology execution layers.", "quality_signal": "official_docs", "retrieval_tags": [ "graphdb", "implementation-evidence", "ontotext", "rdf", "reasoning", "semantic-retrieval", "semantic-search", "shacl", "sparql" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase2-graphrag-032", "title": "LlamaIndex Knowledge Graph RAG documentation", "authors_or_org": "LlamaIndex", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://docs.llamaindex.ai", "doi_or_identifier": null, "venue_or_site": "LlamaIndex Documentation", "abstract_or_summary": "Documentation for building knowledge graph and property graph indexes that can be used with retrieval-augmented generation workflows in LlamaIndex.", "key_claims": [ "Application frameworks expose KG indexes as retrievers for LLM context.", "Graph extraction and graph query components can be embedded into common RAG stacks.", "Developer ergonomics matter for adoption of ontology-like retrieval patterns." ], "ontology_relevance": "Shows how ontology or KG retrieval becomes an application-layer primitive for LLM developers.", "ai_relevance": "Practical framework source for graph indexes and RAG orchestration.", "palantir_relevance": "Useful comparator for lightweight app-framework GraphRAG versus platform-level ontology governance.", "quality_signal": "official_docs", "retrieval_tags": [ "developer-tooling", "knowledge-graph-index", "llamaindex", "property-graph", "rag-framework" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-neo4j-graphrag-docs", "title": "Neo4j GraphRAG documentation and developer guidance", "authors_or_org": "Neo4j", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://neo4j.com/docs/neo4j-graphrag-python/current", "doi_or_identifier": null, "venue_or_site": "Neo4j Docs", "abstract_or_summary": "Neo4j documentation for building retrieval-augmented generation applications using graph data, vector indexes, Cypher retrieval, and graph-aware context expansion.", "key_claims": [ "GraphRAG combines vector retrieval with graph traversal or Cypher retrieval.", "Graph neighborhoods can supply connected context that document chunks alone miss.", "LLM answers can be grounded in graph paths and source nodes.", "GraphRAG can combine vector similarity with explicit graph traversal or Cypher queries.", "Graph neighborhoods provide connected context that chunk retrieval alone can miss.", "LLM answers can be grounded in graph nodes, relationships, and source paths." ], "ontology_relevance": "Demonstrates a property-graph implementation of semantic context and relationship-aware retrieval.", "ai_relevance": "Directly relevant to LLM grounding, retrieval augmentation, and AI answer provenance.", "palantir_relevance": "Contrasts with Palantir AIP by showing GraphRAG as an application pattern over a graph database.", "quality_signal": "official_docs", "retrieval_tags": [ "cypher", "graph-traversal", "graphrag", "llm", "neo4j", "property-graph", "vector-search" ], "triage_tier": "core", "triage_score": 72 }, { "id": "pal-ir-q1-2026-results-2026", "title": "Palantir Reports Q1 2026 U.S. Revenue Growth of 104% Y/Y and Revenue Growth of 85% Y/Y", "authors_or_org": "Palantir Technologies Inc.", "year": 2026, "source_type": "other", "bucket": "palantir", "url": "https://investors.palantir.com/news-details/2026/Palantir-Reports-Q1-2026-U-S--Revenue-Growth-of-104-YY-and-Revenue-Growth-of-85-YY-Raises-FY-2026-Revenue-Guidance-to-71-YY-Growth-and-U-S--Comm-Revenue-Guidance-to-120-YY-Crushing-Consensus-Expectations", "doi_or_identifier": null, "venue_or_site": "Palantir Investor Relations", "abstract_or_summary": "Q1 2026 earnings release reporting continued revenue growth and updated guidance, useful for commercial context around AIP adoption claims.", "key_claims": [ "Palantir reported Q1 2026 U.S. revenue growth of 104% year over year.", "The release raises FY2026 revenue guidance and U.S. commercial guidance.", "The release frames growth around ongoing demand for Palantir software and AI." ], "ontology_relevance": "Indirect context for commercial adoption of the ontology/AIP platform stack.", "ai_relevance": "Investor evidence for current AIP-centered market narrative.", "palantir_relevance": "Primary company-reported growth source; not independent performance proof.", "quality_signal": "primary_source", "retrieval_tags": [ "aip", "investor-relations", "palantir", "q1-2026", "revenue-growth" ], "triage_tier": "core", "triage_score": 72 }, { "id": "pal-ir-q4-2025-results-2026", "title": "Palantir Reports Q4 2025 U.S. Commercial Revenue Growth of 137% Y/Y and Revenue Growth of 70% Y/Y", "authors_or_org": "Palantir Technologies Inc.", "year": 2026, "source_type": "other", "bucket": "palantir", "url": "https://investors.palantir.com/news-details/2026/Palantir-Reports-Q4-2025-U-S--Comm-Revenue-Growth-of-137-YY-and-Revenue-Growth-of-70-YY-Issues-FY-2026-Revenue-Guidance-of-61-YY-and-U-S--Comm-Revenue-Guidance-of-115-YY-Crushing-Consensus-Expectations", "doi_or_identifier": null, "venue_or_site": "Palantir Investor Relations", "abstract_or_summary": "Q4 2025 earnings release and business update presenting Palantir's revenue growth, commercial growth, and AIP product/update narrative.", "key_claims": [ "Q4 2025 U.S. commercial revenue grew 137% year over year according to Palantir.", "Palantir issued FY2026 revenue guidance and highlighted AIP product momentum.", "Investor presentation materials mention Agents and Ontology MCP as new AIP tools.", "Palantir reports rapid U.S. commercial revenue growth and positions AIP as a driver of demand.", "The release is useful for adoption narrative but is not independent technical validation.", "Financial metrics should be separated from architectural claims." ], "ontology_relevance": "Shows how ontology is used in investor/product messaging around AIP scale.", "ai_relevance": "Evidence of Palantir's commercial positioning of AIP and agent tools.", "palantir_relevance": "Primary investor-relations source; claims should be treated as company-reported.", "quality_signal": "primary_source", "retrieval_tags": [ "aip", "commercial-growth", "investor-relations", "ontology-mcp", "palantir", "q4-2025" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase3-pal-sec-10k-2024", "title": "Palantir Technologies Inc. 2024 Form 10-K", "authors_or_org": "Palantir Technologies Inc.; U.S. Securities and Exchange Commission", "year": 2025, "source_type": "other", "bucket": "palantir", "url": "https://www.sec.gov/Archives/edgar/data/1321655/000132165525000022/pltr-20241231.htm", "doi_or_identifier": "sec form 10-k, fiscal year ended 2024-12-31", "venue_or_site": "SEC EDGAR", "abstract_or_summary": "Annual report describing Palantir's principal platforms, including Gotham, Foundry, Apollo, and AIP, and stating that AIP leverages generative AI and LLMs directly within Gotham and Foundry to operationalize AI on enterprise data.", "key_claims": [ "Palantir identifies Gotham, Foundry, Apollo, and AIP as principal platforms.", "Foundry and Gotham transform information into integrated assets reflecting operations.", "AIP uses machine learning and generative AI directly within Gotham and/or Foundry to operationalize AI on enterprise data." ], "ontology_relevance": "SEC filing supports the claim that Palantir's platforms map enterprise information into operational assets.", "ai_relevance": "Primary company filing for AIP as a material platform, with risk-factor context.", "palantir_relevance": "Primary legal/investor source, stronger than marketing for corporate positioning.", "quality_signal": "primary_source", "retrieval_tags": [ "10-k", "aip", "enterprise-ai", "foundry", "gotham", "palantir", "sec" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-poolparty-suite", "title": "PoolParty Semantic Suite product and documentation materials", "authors_or_org": "Semantic Web Company", "year": 2026, "source_type": "docs", "bucket": "commercial", "url": "https://www.poolparty.biz/products/semantic-suite", "doi_or_identifier": null, "venue_or_site": "PoolParty / Semantic Web Company", "abstract_or_summary": "PoolParty materials describe taxonomy management, knowledge graph construction, semantic search, entity extraction, and semantic AI workflows for enterprise content and data.", "key_claims": [ "Controlled vocabularies and taxonomies can provide a pragmatic entry point to enterprise knowledge graphs.", "Entity extraction and tagging link unstructured content to curated concepts.", "Semantic expansion improves search, recommendation, and AI context quality." ], "ontology_relevance": "Useful for SKOS/taxonomy-first ontology adoption and content-oriented semantic enrichment.", "ai_relevance": "Relevant for grounding LLM and RAG workflows in curated enterprise terminology.", "palantir_relevance": "Shows a non-Palantir route from vocabulary governance to semantic AI.", "quality_signal": "official_docs", "retrieval_tags": [ "entity-extraction", "poolparty", "semantic-ai", "semantic-search", "skos", "taxonomy" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase15-rml-specification-heterogeneous-rdf-mapping-2024", "title": "RDF Mapping Language (RML)", "authors_or_org": "RML Community Group", "year": 2024, "source_type": "specification", "bucket": "technical", "url": "https://rml.io/specs/rml", "doi_or_identifier": null, "venue_or_site": "RML.io Specification", "abstract_or_summary": "The RML specification defines RML as a superset of R2RML for customized mapping rules from heterogeneous data structures and serializations to RDF. The specification documents logical sources, reference formulations, term maps, and source-agnostic mapping constructs.", "key_claims": [ "RML expresses mapping rules from heterogeneous data structures to RDF.", "The language extends R2RML while broadening source support beyond relational databases.", "Source-agnostic mapping specifications improve interoperability and repeatability of KG construction pipelines." ], "ontology_relevance": "Makes mapping language semantics concrete for ontology-based data integration and graph construction.", "ai_relevance": "Provides the current technical specification route for agent builders that need to implement heterogeneous source-to-RDF mapping in reproducible pipelines.", "palantir_relevance": "Useful open standard-adjacent comparator for data ingestion and semantic mapping into enterprise ontology layers.", "quality_signal": "official_docs", "retrieval_tags": [ "data-integration", "heterogeneous-data", "knowledge-graph-construction", "mapping-language", "phase15", "rdf", "rml", "semantic-interoperability" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase15-idsa-semantic-interoperability-data-spaces-2024", "title": "Semantic Interoperability in Data Spaces", "authors_or_org": "International Data Spaces Association", "year": 2024, "source_type": "position_paper", "bucket": "technical", "url": "https://zenodo.org/records/10964377", "doi_or_identifier": "10.5281/zenodo.10964377", "venue_or_site": "International Data Spaces Association / Zenodo", "abstract_or_summary": "The IDSA position paper explains why semantic interoperability is required in data spaces and frames common information models, metadata, vocabularies, and governance as necessary for trusted data exchange across participants.", "key_claims": [ "Semantic interoperability in data spaces requires shared information models and understood metadata meanings.", "Data sovereignty and trusted exchange depend on more than transport protocols; participants need shared semantics and governance.", "Data spaces provide a useful lens for cross-organization AI tool and data access governance." ], "ontology_relevance": "Supports the article's claim that ontology-backed AI governance extends beyond one platform into federated ecosystems and data-space semantics.", "ai_relevance": "High-value governance/architecture evidence for AI systems operating across organizations: data meaning, contracts, and usage policies must be machine-readable and shared.", "palantir_relevance": "Useful contrast with centralized operational ontology: data spaces emphasize federation, data sovereignty, and negotiated usage rather than single-vendor integration.", "quality_signal": "official_docs", "retrieval_tags": [ "data-governance", "data-sovereignty", "data-space", "dataspaces", "idsa", "phase15", "semantic-interoperability", "usage-policy" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase16-solid-protocol-linked-data-pods-2025", "title": "Solid Protocol", "authors_or_org": "W3C Solid Community Group", "year": 2025, "source_type": "specification", "bucket": "technical", "url": "https://solidproject.org/TR/protocol", "doi_or_identifier": "solid protocol editor's draft", "venue_or_site": "Solid Project / W3C Community Group", "abstract_or_summary": "The Solid Protocol specifies decentralized data storage through identity, authentication, authorization, linked-data resources, and storage servers called pods. It gives a standards-adjacent model for user-controlled linked data and app interoperability.", "key_claims": [ "Solid separates applications from data storage using linked-data resources and access controls.", "Pods provide a decentralized model for personal or organizational data management.", "Permission-aware agent memory and context access can draw on linked-data pod architectures." ], "ontology_relevance": "Adds a decentralized linked-data governance comparator for identity, storage, and access-controlled semantic resources.", "ai_relevance": "Agentic AI that uses personal or organizational context can benefit from explicit data boundaries, linked-data resources, and permissioned access rather than centralized memory stores.", "palantir_relevance": "Contrasts centralized operational ontology platforms with decentralized linked-data approaches.", "quality_signal": "official_docs", "retrieval_tags": [ "access-control", "decentralized-data", "federated-data-sharing", "linked-data", "ontology-governance", "phase16", "solid", "solid-pods" ], "triage_tier": "core", "triage_score": 72 }, { "id": "p2-llm-ont-034", "title": "SPARQL 1.2 Query Language", "authors_or_org": "World Wide Web Consortium", "year": 2026, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/sparql12-query", "doi_or_identifier": "w3c wd-sparql12-query-20260617", "venue_or_site": "W3C Working Draft", "abstract_or_summary": "Current W3C Working Draft for SPARQL 1.2 Query Language, the standard query language for RDF graph pattern matching and graph retrieval.", "key_claims": [ "Graph query languages are the operational access layer for semantic knowledge bases.", "Standard query semantics enable tool interoperability and verifiable retrieval.", "SPARQL 1.2 indicates ongoing evolution of RDF query capabilities." ], "ontology_relevance": "Operational query layer for ontology-backed RDF stores and semantic data services.", "ai_relevance": "Relevant to LLM tool use where agents query knowledge graphs instead of relying on free-form context.", "palantir_relevance": "Useful contrast to object-set and ontology SDK querying in proprietary platforms.", "quality_signal": "official_docs", "retrieval_tags": [ "agent-tools", "query-language", "rdf", "sparql-1-2", "w3c" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-stardog-virtual-graphs", "title": "Stardog Virtual Graphs documentation", "authors_or_org": "Stardog", "year": 2026, "source_type": "docs", "bucket": "commercial", "url": "https://docs.stardog.com/virtual-graphs", "doi_or_identifier": null, "venue_or_site": "Stardog Docs", "abstract_or_summary": "Documentation for mapping external data sources into Stardog as virtual RDF graphs, enabling SPARQL access and semantic integration without fully copying source data.", "key_claims": [ "Virtual graphs expose external enterprise data through semantic mappings.", "Ontology-backed graph access can decouple business queries from physical schemas.", "Virtualization can be combined with materialized graph data when needed." ], "ontology_relevance": "Shows how enterprise ontology can act as a semantic mapping layer over operational data.", "ai_relevance": "Provides governed graph context that can feed search, analytics, or GraphRAG systems.", "palantir_relevance": "Comparable to Palantir-style ontology over source systems, but implemented with RDF/SPARQL semantic virtualization.", "quality_signal": "official_docs", "retrieval_tags": [ "data-virtualization", "enterprise-kg", "implementation-evidence", "rdf", "semantic-layer", "sparql", "stardog", "virtual-graph", "virtual-graphs" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase15-gaiax-ontology-compliance-policy-reasoning-2023", "title": "The Role of Ontologies in Gaia-X", "authors_or_org": "Gaia-X European Association for Data and Cloud AISBL", "year": 2023, "source_type": "technical_article", "bucket": "technical", "url": "https://gaia-x.eu/the-role-of-ontologies-in-gaia-x", "doi_or_identifier": null, "venue_or_site": "Gaia-X", "abstract_or_summary": "Gaia-X documents the role of ontologies in compliance and policy reasoning. Gaia-X uses RDF triples, verifiable credentials, SHACL shapes, JSON-LD schemas, and OWL ontology artifacts to check consistency and support machine-readable trust and compliance operations.", "key_claims": [ "Gaia-X uses ontologies for compliance and policy reasoning.", "RDF triples, SHACL shapes, JSON-LD schemas, OWL ontology artifacts, and verifiable credentials support machine-readable trust operations.", "Federated ecosystems need shared semantic descriptions to verify claims and maintain interoperability." ], "ontology_relevance": "Strongly supports the governance route: ontology can power compliance checking and policy reasoning, not merely descriptive metadata.", "ai_relevance": "Gaia-X is a concrete current example of ontology, SHACL, JSON-LD, and verifiable credentials used for automated compliance and policy reasoning in federated digital ecosystems.", "palantir_relevance": "Useful comparator for Palantir governance claims because Gaia-X exposes an open/federated ontology and credential-based compliance model.", "quality_signal": "official_docs", "retrieval_tags": [ "compliance", "data-space", "gaia-x", "json-ld", "ontology-governance", "phase15", "policy-reasoning", "shacl", "verifiable-credentials" ], "triage_tier": "core", "triage_score": 72 }, { "id": "industry-commercial-topquadrant-edg", "title": "TopBraid EDG enterprise knowledge graph and data governance materials", "authors_or_org": "TopQuadrant", "year": 2026, "source_type": "docs", "bucket": "commercial", "url": "https://www.topquadrant.com/products/topbraid-edg", "doi_or_identifier": null, "venue_or_site": "TopQuadrant", "abstract_or_summary": "Product and documentation material for TopBraid EDG, covering enterprise ontology, taxonomy, reference data, data catalog, governance workflows, and SHACL-based validation.", "key_claims": [ "Ontology and taxonomy assets need governance workflows, stewardship, versioning, and publishing.", "Business glossaries, reference data, taxonomies, and catalogs can be managed as connected semantic assets.", "SHACL constraints can operationalize semantic governance." ], "ontology_relevance": "Shows ontology as a governed enterprise asset lifecycle, not just a schema file.", "ai_relevance": "Governed semantic assets can supply controlled vocabulary and policy context to AI systems.", "palantir_relevance": "Comparable to Palantir's governed object model, but centered on Semantic Web standards and governance workflows.", "quality_signal": "official_docs", "retrieval_tags": [ "data-governance", "ontology-management", "shacl", "taxonomy", "topbraid-edg", "topquadrant" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase15-yarrrml-human-readable-rdf-generation-rules-2025", "title": "YARRRML: Human Readable Text-Based Representation for Declarative Generation Rules", "authors_or_org": "RML Community Group", "year": 2025, "source_type": "specification", "bucket": "technical", "url": "https://w3id.org/yarrrml/spec", "doi_or_identifier": null, "venue_or_site": "YARRRML Specification", "abstract_or_summary": "YARRRML is a human-readable YAML-based representation for declarative generation rules that can be converted to RML. It lowers authoring complexity for users defining RDF generation rules over heterogeneous sources.", "key_claims": [ "YARRRML provides a human-readable representation for declarative RDF generation rules.", "The format can simplify authoring mappings over multiple existing data sources.", "Usable mapping syntax helps make semantic integration maintainable by data teams, not only semantic-web experts." ], "ontology_relevance": "Adds a usability layer to ontology mappings: maintainable semantic integration requires reviewable mapping artifacts.", "ai_relevance": "Practical agent and data-engineering systems need mapping formats humans can review and maintain; YARRRML is evidence that mapping usability matters for ontology-backed AI pipelines.", "palantir_relevance": "Relevant comparator for enterprise ontology builders because business-facing mapping workflows need readable, inspectable rules.", "quality_signal": "official_docs", "retrieval_tags": [ "data-integration", "knowledge-graph-construction", "mapping-language", "phase15", "rdf", "rml", "semantic-interoperability", "yaml", "yarrrml" ], "triage_tier": "core", "triage_score": 72 }, { "id": "phase2-hu-2024-graphrag-survey", "title": "A Survey of Graph Retrieval-Augmented Generation for Customized Large Language Models", "authors_or_org": "Yuntong Hu; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2501.13958", "doi_or_identifier": "arxiv:2501.13958", "venue_or_site": "arXiv", "abstract_or_summary": "Survey of GraphRAG methods for customized LLMs, covering graph construction, graph retrieval, augmentation strategies, applications, and evaluation.", "key_claims": [ "GraphRAG methods differ in how they construct graphs, retrieve subgraphs, augment prompts, and evaluate answers.", "Customized LLM systems need domain-specific graph construction and maintenance strategies.", "Ontology can supply schema, constraints, and provenance that many graph-RAG pipelines otherwise generate weakly.", "GraphRAG customizes LLMs by externalizing domain structure into graph memory.", "Graph construction, retrieval strategy, and update mechanisms are central design variables.", "GraphRAG needs better benchmarks for factuality, relevance, and domain transfer." ], "ontology_relevance": "Helps map GraphRAG technical variants to ontology engineering choices.", "ai_relevance": "Recent survey source for GraphRAG architectures and applications.", "palantir_relevance": "Useful comparator for Palantir's ontology-augmented generation and operational graph retrieval.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "customized-llm", "domain-adaptation", "evaluation", "graph-memory", "graphrag", "knowledge-graph", "rag", "retrieval", "survey" ], "triage_tier": "core", "triage_score": 70 }, { "id": "phase2-sun-2025-agentic-rag-survey", "title": "Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG", "authors_or_org": "Survey authors", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2501.09136", "doi_or_identifier": "arxiv:2501.09136", "venue_or_site": "arXiv", "abstract_or_summary": "Survey of agentic RAG systems that use agents for planning, query decomposition, tool use, memory, retrieval orchestration, and answer synthesis.", "key_claims": [ "Agentic RAG extends RAG with planning, tool use, memory, and dynamic retrieval strategies.", "Evaluation should cover both final answers and intermediate retrieval/tool decisions.", "Ontology and knowledge graphs can define retrievable objects, permissible actions, and provenance for agentic RAG." ], "ontology_relevance": "Useful for connecting ontology indexes and graph relations to agentic retrieval workflows.", "ai_relevance": "Recent survey on agentic RAG architectures and evaluation challenges.", "palantir_relevance": "Strong comparator to Palantir AIP agents using ontology resources and controlled actions.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "agent-memory", "agentic-rag", "agents", "evaluation", "knowledge-graph", "rag", "tool-use" ], "triage_tier": "core", "triage_score": 70 }, { "id": "phase14-croissant-ml-ready-dataset-metadata-2024", "title": "Croissant: A Metadata Format for ML-Ready Datasets", "authors_or_org": "Mubashara Akhtar; Omar Benjelloun; Costanza Conforti; Luca Foschini; Pieter Gijsbers; Joan Giner-Miguelez; Sujata Goswami; Nitisha Jain; Michalis Karamousadakis; Michael Kuchnik; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3650203.3663326", "doi_or_identifier": "10.1145/3650203.3663326", "venue_or_site": "ACM DEEM", "abstract_or_summary": "Croissant is a metadata format for ML-ready datasets developed by an MLCommons community and published in the 2024 ACM DEEM workshop. It builds on schema.org/JSON-LD to describe dataset metadata, resources, structure, and ML semantics so datasets can be discovered, loaded, and reused across tools and repositories.", "key_claims": [ "Croissant creates a shared metadata representation for ML datasets across tools, frameworks, and platforms.", "The format improves discoverability, portability, interoperability, reproducibility, and responsible-use metadata for datasets.", "Croissant builds on schema.org/JSON-LD and adds ML-specific semantics for resources, record sets, and fields." ], "ontology_relevance": "Supports the argument that AI-ready knowledge infrastructure requires dataset-level semantics, field structure, provenance, and machine-readable contracts.", "ai_relevance": "High-value evidence that responsible and portable ML depends on shared machine-readable dataset metadata, not only model cards or free-text dataset descriptions.", "palantir_relevance": "Useful comparator for Foundry-style dataset and ontology metadata: Croissant shows an open metadata route for ML data portability and reuse.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "croissant", "dataset-metadata", "json-ld", "machine-learning-data", "ml-ready-data", "phase14", "responsible-ai", "schema-org", "semantic-interoperability" ], "triage_tier": "core", "triage_score": 70 }, { "id": "oa-https-doi-org-10-36227-techrxiv-175624549-98427022-v1", "title": "Enterprise Digital Twins in Financial Services: Convergence of Architecture, Operations, and Engineering", "authors_or_org": "Glen Stokes", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4413638561", "doi_or_identifier": "10.36227/techrxiv.175624549.98427022/v1", "venue_or_site": "", "abstract_or_summary": "The Enterprise Digital Twin (EDT) represents the convergence of decades-long parallel innovations in enterprise architecture (Zachman, TOGAF, ArchiMate), systems engineering (Agile, DevOps), and operational management (ITIL, SRE, observability). This paper demonstrates that EDTs are not an abrupt invention but the culmination of six decades of theory and practice-from 1960s socio-technical systems to contemporary AI-enabled, graph-driven models. Recent advances in graph databases, semantic modeling, and AI-augmented engineering now make implementation feasible, with EDTs providing the context that is so important for agentic AI. The paper examines EDT foundations in ontology, taxonomy, and graph theory; reviews enabling standards and technologies; outlines socio-technical implementation methodology; and analyzes real-world applications including ING's capability transformation, Lloyds' observability model, and Singapore's national digital twin. FSIs can adopt EDTs with confidence, applying proven frameworks through modern technologies. The purpose of an EDT is to help FSIs to address one of their key challenges-accelerating innovation without compromising compliance and security-but the principles apply equally to healthcare and government. Starting to build an EDT presents an opportunity for architecture, engineering and operations teams to pool their strengths around a common business problem. We discuss the rise of an Autonomous Agentic Twin: orchestrated agentic swarms that cooperate with the EDT to autonomously execute business processes, make governance-compliant decisions, and adapt to changing conditions. This evolution represents the rapidly approaching future of autonomous enterprises.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "core", "triage_score": 70 }, { "id": "oa-https-doi-org-10-3390-electronics14183583", "title": "ForestGPT and Beyond: A Trustworthy Domain-Specific Large Language Model Paving the Way to Forestry 5.0", "authors_or_org": "Florian Sommer, Benno Eberhard, Andreas Holzinger", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4414098962", "doi_or_identifier": "10.3390/electronics14183583", "venue_or_site": "Electronics", "abstract_or_summary": "Large language models (LLMs) such as Chat Generative Pre-Trained Transformer (ChatGPT) are increasingly used across domains, yet their generic training data and propensity for hallucination limit reliability in safety-critical fields like forestry. This paper outlines the conception and prototype of ForestGPT, a domain-specialised assistant designed to support forest professionals while preserving expert oversight. It addresses two looming risks: unverified adoption of generic outputs and professional mistrust of opaque algorithms. We propose a four-level development path: (1) pre-training a transformer on curated forestry literature to create a baseline conversational tool; (2) augmenting it with Retrieval-Augmented Generation to ground answers in local and time-sensitive documents; (3) coupling growth simulators for scenario modeling; and (4) integrating continuous streams from sensors, drones and machinery for real-time decision support. A Level-1 prototype, deployed at Futa Expo 2025 via a mobile app, successfully guided multilingual visitors and demonstrated the feasibility of lightweight fine-tuning on open-weight checkpoints. We analyse technical challenges, multimodal grounding, continual learning, safety certification, and social barriers including data sovereignty, bias and change management. Results indicate that trustworthy, explainable, and accessible LLMs can accelerate the transition to Forestry 5.0, provided that human-in-the-loop guardrails remain central. Future work will extend ForestGPT with full RAG pipelines, simulator coupling and autonomous data ingestion. Whilst exemplified in forestry, a complex, safety-critical, and ecologically vital domain, the proposed architecture and development path are broadly transferable to other sectors that demand trustworthy, domain-specific language models under expert oversight.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "core", "triage_score": 70 }, { "id": "industry-commercial-neo4j-customer-stories", "title": "Neo4j customer stories for enterprise graph and knowledge graph applications", "authors_or_org": "Neo4j", "year": 2026, "source_type": "case_study", "bucket": "commercial", "url": "https://neo4j.com/customers", "doi_or_identifier": null, "venue_or_site": "Neo4j Customer Stories", "abstract_or_summary": "Neo4j customer stories cover graph applications in recommendations, fraud, master data, life sciences, supply chain, identity, and knowledge discovery.", "key_claims": [ "Property graphs are useful where business value depends on relationship traversal and pattern discovery.", "Connected data models can power operational applications and analytics.", "Customer stories demonstrate adoption breadth but often need independent validation for quantitative claims." ], "ontology_relevance": "Useful for property-graph alternatives to formal RDF/OWL ontology in enterprise practice.", "ai_relevance": "Graph data science and GraphRAG patterns can build on these connected domain models.", "palantir_relevance": "Provides non-Palantir examples of operational graphs used by enterprise applications.", "quality_signal": "primary_source", "retrieval_tags": [ "customer-stories", "graph-analytics", "knowledge-graph", "neo4j", "property-graph" ], "triage_tier": "core", "triage_score": 70 }, { "id": "nhs-fdp-privacy-notice-2024", "title": "NHS Federated Data Platform privacy notice", "authors_or_org": "NHS England", "year": 2024, "source_type": "webpage", "bucket": "technical", "url": "https://www.england.nhs.uk/digitaltechnology/nhs-federated-data-platform/security-privacy/nhs-fdp-privacy-notice", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "Official privacy notice describing processing of personal data in the NHS Federated Data Platform, including purposes, data sources, access, protection, retention, and legal grounds.", "key_claims": [ "The FDP processes personal data for specified NHS operational purposes.", "The notice describes who is responsible for processing and who has access.", "The notice presents safeguards, legal bases, and rights information.", "NHS England describes the categories and purposes of data processing in the FDP.", "The notice is central for assessing public claims about data use limits and legal basis.", "It provides a primary governance source to compare against campaigner and parliamentary concerns." ], "ontology_relevance": "Important for governance analysis of a healthcare operational data/ontology platform.", "ai_relevance": "Relevant to AI governance where patient data becomes operational context for analysis and automation.", "palantir_relevance": "Official public-sector privacy context for Palantir Foundry deployment.", "quality_signal": "primary_source", "retrieval_tags": [ "data-governance", "fdp", "governance", "health-data", "nhs", "palantir", "personal-data", "privacy-notice" ], "triage_tier": "core", "triage_score": 70 }, { "id": "phase14-xiao-2018-ontology-based-data-access-survey", "title": "Ontology-Based Data Access: A Survey", "authors_or_org": "Guohui Xiao; Diego Calvanese; Roman Kontchakov; Domenico Lembo; Antonella Poggi; Riccardo Rosati; Michael Zakharyaschev", "year": 2018, "source_type": "paper", "bucket": "academic", "url": "https://www.ijcai.org/proceedings/2018/0777.pdf", "doi_or_identifier": "10.24963/ijcai.2018/777", "venue_or_site": "IJCAI", "abstract_or_summary": "The IJCAI survey presents ontology-based data access (OBDA) as a semantic paradigm for user-friendly access to data repositories, especially relational data. It covers the ingredients of OBDA, theoretical results, techniques, applications, and future challenges.", "key_claims": [ "OBDA provides a semantic access layer where users query data repositories through an ontology-mediated conceptual view.", "Mappings connect the ontology layer to underlying relational sources, allowing query rewriting and data access without full consolidation.", "OBDA research connects ontology, databases, query answering, mappings, and heterogeneous data integration." ], "ontology_relevance": "Directly supports the semantic-layer route: ontology can mediate access to existing databases and operational systems rather than requiring all data to be copied into one graph store.", "ai_relevance": "OBDA and virtual knowledge graphs are key for AI systems that must query governed enterprise data through a conceptual layer without physically consolidating all data.", "palantir_relevance": "A strong academic comparator for Palantir-style semantic access over heterogeneous enterprise data sources.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "data-integration", "obda", "ontology-based-data-access", "phase14", "query-rewriting", "semantic-layer", "virtual-knowledge-graph" ], "triage_tier": "core", "triage_score": 70 }, { "id": "phase15-dimou-2014-rml-integrated-rdf-mappings", "title": "RML: A Generic Language for Integrated RDF Mappings of Heterogeneous Data", "authors_or_org": "Anastasia Dimou; Miel Vander Sande; Pieter Colpaert; Ruben Verborgh; Erik Mannens; Rik Van de Walle", "year": 2014, "source_type": "paper", "bucket": "academic", "url": "https://ceur-ws.org/Vol-1184/ldow2014_paper_01.pdf", "doi_or_identifier": null, "venue_or_site": "LDOW 2014", "abstract_or_summary": "RML extends R2RML from relational databases to heterogeneous sources and serializations. The LDOW paper introduces RML as a generic mapping language for integrated RDF mappings over CSV, XML, JSON, and other data sources.", "key_claims": [ "RML generalizes R2RML to heterogeneous data sources and serializations.", "Declarative mappings can make heterogeneous data integration into RDF more reusable and less tool-specific.", "RML exposes mapping logic that would otherwise remain hidden in extraction code." ], "ontology_relevance": "Strengthens the data-to-ontology pipeline: mappings are first-class artifacts that connect source formats to ontology terms and KG structures.", "ai_relevance": "RML is important for AI knowledge infrastructure because enterprise and scientific data rarely live only in relational databases; agents need provenance-preserving mappings from many formats into a graph layer.", "palantir_relevance": "A strong open comparator for turning heterogeneous enterprise data into semantic objects without losing mapping provenance.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "data-integration", "heterogeneous-data", "knowledge-graph-construction", "mapping-language", "phase15", "rdf", "rml", "semantic-interoperability" ], "triage_tier": "core", "triage_score": 70 }, { "id": "phase2-ali-2024-trustworthy-graphrag", "title": "Trustworthy GraphRAG: A Survey", "authors_or_org": "Survey authors", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2501.02157", "doi_or_identifier": "arxiv:2501.02157", "venue_or_site": "arXiv", "abstract_or_summary": "Survey-style preprint synthesizing trustworthiness dimensions for GraphRAG, including reliability, explainability, privacy, robustness, fairness, and evaluation challenges.", "key_claims": [ "GraphRAG trustworthiness depends on graph construction quality, retrieval reliability, explanation, privacy, robustness, and evaluation.", "Graph structure can improve traceability but also introduces new failure modes from extraction, entity resolution, and stale edges.", "Ontology-grounded GraphRAG should be evaluated as a graph lifecycle plus generation system." ], "ontology_relevance": "Useful bridge between ontology quality, graph construction, and trustworthy GraphRAG evaluation.", "ai_relevance": "Directly relevant to GraphRAG reliability and governance literature.", "palantir_relevance": "Comparable to Palantir ontology-grounded AI where graph context and action tools must be trusted.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "evaluation", "graphrag", "knowledge-graph", "privacy", "provenance", "robustness", "trustworthy-ai" ], "triage_tier": "core", "triage_score": 70 } ], "candidate": [ { "id": "phase3-pal-aip-evals-results-dataset-2026", "title": "AIP Evals: Write run results to a dataset", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/aip-evals/results-dataset", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation explaining how AIP Evals run results can be written to a dataset, including function outputs, evaluator results, metadata, and errors.", "key_claims": [ "AIP Evals can write generated run information to a configured dataset.", "Stored information can include function outputs, evaluator results, metadata, and errors.", "For ontology-editing functions, tested functions may not produce a normal function output." ], "ontology_relevance": "Shows how evaluation evidence can itself become data available elsewhere in the platform.", "ai_relevance": "Supports auditability and review workflows for AI evaluations.", "palantir_relevance": "Primary source for surfacing AIP eval results to nontechnical reviewers.", "quality_signal": "official_docs", "retrieval_tags": [ "aip-evals", "auditability", "metadata", "palantir", "results-dataset" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase3-pal-functions-edits-overview-2026", "title": "Functions: Ontology edits overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/edits-overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official overview defining an ontology edit as creating, modifying, or deleting an object, and explaining function-backed actions for operational use.", "key_claims": [ "An Ontology edit is the act of creating, modifying, or deleting an object.", "Functions can return Ontology edits for use in a function-backed action.", "Operational edits are routed through Actions rather than arbitrary code execution alone." ], "ontology_relevance": "Documents the kinetic side of Palantir's ontology model.", "ai_relevance": "Important for evaluating write-capable AI agents and their action boundaries.", "palantir_relevance": "Primary source for object writeback semantics.", "quality_signal": "official_docs", "retrieval_tags": [ "action-layer", "function-backed-actions", "functions", "ontology-edits", "palantir", "writeback" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase4-palantir-functions-objects-getting-started-2026", "title": "Getting started with functions on objects", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/foo-getting-started", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official guide showing how functions operate on ontology objects, supporting Palantir's claim that Ontology is an executable application substrate rather than only a semantic model.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "developer-workflow", "functions", "object-types", "ontology", "palantir" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "pal-doc-object-edits-2026", "title": "Object edits and materializations: How user edits are applied", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/object-edits/how-edits-applied", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Technical documentation for how ontology Actions apply user edits, including object database indexing, Funnel-managed queues and datasets, version checks, persistent storage, and conflict resolution.", "key_claims": [ "Actions apply data-modification logic to object database indexes and persist edits through Foundry datasets managed by Funnel.", "Object Storage V2 tracks offsets and supports schema migration for user edits.", "Conflict resolution can apply user edits by default or use a most-recent-value strategy.", "Actions apply modification logic to object database indexes and persist edits through Foundry datasets managed by Funnel." ], "ontology_relevance": "Shows that Palantir's ontology is writable and has operational state semantics, not just static metadata.", "ai_relevance": "Critical for understanding how AI-mediated actions can be tested, governed, and written back to operational systems.", "palantir_relevance": "Primary technical source for writeback/action semantics.", "quality_signal": "official_docs", "retrieval_tags": [ "actions", "conflict-resolution", "funnel", "object-edits", "object-storage", "object-storage-v2", "palantir", "writeback" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase2-pal-object-materializations-2026", "title": "Object edits and materializations: Materializations", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/object-edits/materializations", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation explaining materializations of ontology data that combine indexed data with user edits for downstream pipelines or downloads.", "key_claims": [ "Materializations provide latest object state by combining input datasource data and user edits.", "Object Storage V2 makes materializations optional for enabling edits but useful for downstream pipelines and downloads.", "Ontology writeback creates a need to reconcile source data, edited state, and downstream consumption." ], "ontology_relevance": "Important for understanding operational ontology state and writeback-derived datasets.", "ai_relevance": "Relevant to AI workflows that need current state, user edits, and downstream pipeline integration.", "palantir_relevance": "Primary source for Palantir materialized object state behavior.", "quality_signal": "official_docs", "retrieval_tags": [ "downstream-pipeline", "materialization", "object-edits", "object-storage", "palantir", "writeback" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase2-pal-ontology-mcp-debugging-2026", "title": "Ontology MCP: Debugging with MCP Inspector", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-mcp/debugging-with-mcp-inspector", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation on using MCP Inspector to list tools exposed by an Ontology MCP server, run tools, inspect success/error responses, and verify configured restrictions.", "key_claims": [ "Ontology MCP tool lists should match ontology resources configured in application restrictions.", "Tool execution responses and permission errors can be inspected during debugging.", "Currently, Ontology MCP focuses on tools and does not expose prompts or resources in MCP Inspector." ], "ontology_relevance": "Provides concrete evidence of ontology resources being debugged as tool interfaces.", "ai_relevance": "Relevant to production agent tool-chain observability and debugging.", "palantir_relevance": "Useful primary source for operationalizing and inspecting Palantir OMCP tool boundaries.", "quality_signal": "official_docs", "retrieval_tags": [ "debugging", "mcp-inspector", "ontology-mcp", "palantir", "permissions", "tool-use" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase2-pal-ontology-mcp-example-workflows-2026", "title": "Ontology MCP: Example MCP workflows", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-mcp/example-mcp-workflows", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation showing example use cases where Ontology MCP connects desktop agents such as Claude.ai, Microsoft Copilot Studio, and Gemini Enterprise to organizational ontology resources.", "key_claims": [ "Ontology MCP can connect desktop agents to an organization's source of truth.", "External agents can read and write data and execute actions governed by the ontology.", "The examples show Palantir's strategy of making ontology resources available outside the native Foundry UI." ], "ontology_relevance": "Concrete evidence of ontology as an agent-accessible source-of-truth interface.", "ai_relevance": "Important source for external agent integration with governed enterprise data/actions.", "palantir_relevance": "Primary source for Palantir's OMCP use cases with external AI agents.", "quality_signal": "official_docs", "retrieval_tags": [ "desktop-agents", "governed-action", "mcp", "ontology-mcp", "palantir", "source-of-truth" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase3-pal-ontology-mcp-overview-2026", "title": "Ontology MCP: Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://www.palantir.com/docs/foundry/ontology-mcp/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official documentation for Ontology MCP, a Developer Console feature that exposes application ontology resources as Model Context Protocol tools.", "key_claims": [ "Ontology MCP exposes ontology resources as MCP tools.", "AI agents and external systems can interact with the ontology as MCP clients.", "The feature bridges Developer Console applications and external agent ecosystems." ], "ontology_relevance": "Shows Palantir translating ontology resources into agent-discoverable tools.", "ai_relevance": "Direct evidence of ontology-grounded AI agent interoperability through MCP.", "palantir_relevance": "High-value official source for Palantir's MCP strategy.", "quality_signal": "official_docs", "retrieval_tags": [ "agents", "developer-console", "mcp", "model-context-protocol", "ontology-mcp", "palantir" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase3-pal-osdk-overview-2026", "title": "Ontology SDK (OSDK): Overview", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-sdk/overview", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official OSDK overview explaining that developers can access the Ontology from TypeScript, Python, Java, or OpenAPI-based clients.", "key_claims": [ "The OSDK exposes the full power of the Ontology from developer environments.", "It supports TypeScript, Python, Java, and OpenAPI specs for other languages.", "Client code is generated against ontology resources." ], "ontology_relevance": "Shows the ontology as a programmable API surface.", "ai_relevance": "AI applications and agent tools can be built against typed ontology clients.", "palantir_relevance": "Core evidence for developer access to ontology objects, actions, and functions.", "quality_signal": "official_docs", "retrieval_tags": [ "developer-console", "ontology-sdk", "openapi", "osdk", "palantir" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase4-palantir-python-functions-objects-2026", "title": "Python: Functions on objects", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/python-functions-on-objects", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official Python documentation for using functions with ontology objects, indicating that ontology-backed logic can be implemented in familiar programming environments.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "developer-api", "functions", "ontology", "palantir", "python" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase3-pal-osdk-subscriptions-2026", "title": "Subscribe to Ontology changes with the TypeScript OSDK", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/ontology-sdk/typescript-subscriptions", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official page documenting real-time subscriptions to ontology object changes using the TypeScript OSDK.", "key_claims": [ "TypeScript OSDK can subscribe to real-time updates about changes to ontology objects.", "Object-set changes can be streamed to clients.", "Operational applications can react to ontology state changes." ], "ontology_relevance": "Adds dynamic/evented behavior to the ontology-as-operational-layer thesis.", "ai_relevance": "Relevant to AI agents and apps that monitor changing operational state.", "palantir_relevance": "Technical evidence for real-time ontology-backed applications.", "quality_signal": "official_docs", "retrieval_tags": [ "ontology-changes", "osdk", "palantir", "real-time", "subscriptions" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "phase4-palantir-typescript-v2-ontology-edits-2026", "title": "TypeScript v2: Ontology edits", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "docs", "bucket": "palantir", "url": "https://palantir.com/docs/foundry/functions/typescript-v2-ontology-edits", "doi_or_identifier": null, "venue_or_site": "Palantir Foundry Documentation", "abstract_or_summary": "Official TypeScript guidance for writing ontology edits, showing developer-level interfaces for mutating operational objects and links.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "developer-api", "functions", "ontology-edits", "palantir", "typescript" ], "triage_tier": "candidate", "triage_score": 68 }, { "id": "industry-commercial-aws-neptune-docs", "title": "Amazon Neptune documentation", "authors_or_org": "Amazon Web Services", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://docs.aws.amazon.com/neptune", "doi_or_identifier": null, "venue_or_site": "AWS Documentation", "abstract_or_summary": "Official documentation for Amazon Neptune graph database and graph analytics services, including RDF/SPARQL, property graph APIs, graph analytics, and integration with AWS services.", "key_claims": [ "Managed graph infrastructure can support RDF and property graph workloads.", "Enterprise KG systems usually require surrounding ingestion, analytics, search, security, and application layers.", "Graph ML and analytics can operate over connected enterprise data.", "Managed graph infrastructure can support both RDF and property graph workloads.", "Enterprise KG systems require ingestion, query, analytics, search, security, and application layers.", "Cloud graph services can become retrieval substrates for LLM systems." ], "ontology_relevance": "Shows the infrastructure substrate for ontology-backed and graph-backed applications in cloud architectures.", "ai_relevance": "Neptune can support graph analytics, ML, and GraphRAG-style applications when paired with AI services.", "palantir_relevance": "Useful contrast to Palantir's integrated platform: AWS supplies composable graph infrastructure rather than one ontology operating layer.", "quality_signal": "official_docs", "retrieval_tags": [ "amazon-neptune", "aws", "cloud-kg", "graph-analytics", "property-graph", "rdf", "sparql" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase4-palantir-fdp-evaluation-partner-tender-2026", "title": "Federated Data Platform (FDP) - Evaluation Partner", "authors_or_org": "Crown Commercial Service", "year": 2026, "source_type": "government_record", "bucket": "palantir", "url": "https://www.find-tender.service.gov.uk/procurement/ocds-h6vhtk-05ec4c", "doi_or_identifier": null, "venue_or_site": "Find a Tender Service", "abstract_or_summary": "UK procurement notice for an evaluation partner for the NHS Federated Data Platform, indicating continuing official interest in evaluating FDP implementation.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "evaluation", "fdp", "governance", "nhs", "palantir", "procurement" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase4-palantir-nhs-fdp-ig-cyber-2023", "title": "NHS Federated Data Platform and Associated Services Document 2: Information Governance and Cyber Security", "authors_or_org": "NHS England", "year": 2023, "source_type": "government_record", "bucket": "palantir", "url": "https://atamis-1928.my.salesforce-sites.com/servlet/servlet.FileDownload?file=00P8d000009dbdlEAA", "doi_or_identifier": null, "venue_or_site": "NHS England / Atamis procurement portal", "abstract_or_summary": "NHS procurement document focused on information governance and cyber security expectations for the Federated Data Platform.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "cybersecurity", "fdp", "information-governance", "nhs", "palantir", "procurement" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase4-palantir-hansard-fdp-contract-award-2023", "title": "NHS Federated Data Platform: Contract Award", "authors_or_org": "UK Parliament", "year": 2023, "source_type": "government_record", "bucket": "palantir", "url": "https://hansard.parliament.uk/commons/2023-11-21/debates/23112160000011/NHSFederatedDataPlatformContractAward", "doi_or_identifier": null, "venue_or_site": "Hansard, UK Parliament", "abstract_or_summary": "Parliamentary record of questions and ministerial responses following the NHS Federated Data Platform contract award to a Palantir-led consortium.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "governance", "nhs", "palantir", "parliament", "procurement" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase16-noy-klein-2004-ontology-evolution-schema-evolution", "title": "Ontology Evolution: Not the Same as Schema Evolution", "authors_or_org": "Natalya F. Noy; Michel Klein", "year": 2004, "source_type": "paper", "bucket": "academic", "url": "https://www.sciencedirect.com/science/article/pii/S1570826804000224", "doi_or_identifier": null, "venue_or_site": "Web Semantics: Science, Services and Agents on the World Wide Web", "abstract_or_summary": "Classic ontology evolution paper distinguishing ontology evolution from database schema evolution. It frames ontology change as a semantic and collaborative process where meaning, usage, dependent applications, and mappings must be managed as the ontology changes.", "key_claims": [ "Ontology evolution is not the same as database schema evolution because ontology changes alter shared meanings and commitments.", "Versioning, change propagation, dependent artifacts, and collaborative negotiation are central to ontology maintenance.", "AI systems using ontology-backed memory or tools need explicit handling of semantic drift and change impact.", "DOI should be rechecked from the publisher page before formal submission." ], "ontology_relevance": "Foundational source for ontology evolution, change management, versioning, and semantic drift.", "ai_relevance": "Agentic AI systems that depend on ontology memory need versioning, migration, provenance, and change impact analysis rather than silent prompt-level updates.", "palantir_relevance": "Useful for analyzing operational ontology updates, object-model migration, and downstream workflow dependence in enterprise platforms.", "quality_signal": "widely_cited", "retrieval_tags": [ "change-management", "ontology-evolution", "ontology-governance", "ontology-maintenance", "ontology-versioning", "phase16", "schema-evolution", "semantic-drift" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase6-odcs-310-open-data-contract-standard", "title": "Open Data Contract Standard v3.1.0", "authors_or_org": "Bitol / LF AI & Data", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://bitol-io.github.io/open-data-contract-standard/v3.1.0", "doi_or_identifier": "odcs v3.1.0", "venue_or_site": "Open Data Contract Standard Documentation", "abstract_or_summary": "Defines YAML keys and sections for data contracts, including schema, quality, roles, service levels, support, pricing, and governance metadata.", "key_claims": [ "Data contracts encode producer-consumer expectations in versionable YAML.", "Contracts combine structural schema with quality and operational commitments.", "The specification is designed for automation and validation." ], "ontology_relevance": "Can map contract fields into ontology classes for datasets, owners, quality rules, and obligations.", "ai_relevance": "Gives agents explicit guardrails for which data is valid, owned, supported, and fit for use.", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "data-contract", "data-governance", "odcs", "phase6", "producer-consumer", "quality-rules", "schema", "validation" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase4-palantir-sec-proxy-2026", "title": "Palantir Technologies Inc. 2026 Proxy Statement", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "sec_filing", "bucket": "palantir", "url": "https://www.sec.gov/Archives/edgar/data/1321655/000132165526000019/pltr-20260423.htm", "doi_or_identifier": null, "venue_or_site": "U.S. Securities and Exchange Commission EDGAR", "abstract_or_summary": "SEC proxy statement for Palantir, useful for governance, executive compensation, shareholder structure, and board oversight context around the company's AI and public-sector business.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "governance", "investor-disclosure", "palantir", "proxy", "sec" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase3-pal-ir-q1-2026-10q", "title": "Palantir Technologies Inc. Q1 2026 Form 10-Q", "authors_or_org": "Palantir Technologies Inc.; U.S. Securities and Exchange Commission", "year": 2026, "source_type": "other", "bucket": "palantir", "url": "https://investors.palantir.com/files/2026%20Q1%20PLTR%2010-Q.pdf", "doi_or_identifier": "sec form 10-q, quarter ended 2026-03-31", "venue_or_site": "Palantir Investor Relations / SEC filing PDF", "abstract_or_summary": "Quarterly filing stating that the Ontology has evolved as the heart of Palantir's platforms by activating data and analytics inside operations and connecting data, analytics, operational teams, and AI.", "key_claims": [ "The Ontology is described as the heart of Palantir's platforms.", "Palantir says it activates data and analytics inside operations.", "The filing connects the Ontology with real-time connectivity between data, analytics, operational teams, and AI." ], "ontology_relevance": "Strong primary-source corporate claim for ontology as the central operational layer.", "ai_relevance": "Links ontology explicitly to AI in Palantir's public investor disclosures.", "palantir_relevance": "SEC/investor material useful for weighting product claims against business strategy.", "quality_signal": "primary_source", "retrieval_tags": [ "10-q", "ai", "investor-relations", "ontology", "operations", "palantir" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "p2-llm-ont-035", "title": "SHACL 1.2 Core", "authors_or_org": "World Wide Web Consortium", "year": 2026, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/shacl12-core", "doi_or_identifier": "w3c wd-shacl12-core-20260622", "venue_or_site": "W3C Working Draft", "abstract_or_summary": "Current W3C Working Draft for SHACL 1.2 Core, describing constraints and validation for RDF graphs using shapes.", "key_claims": [ "Graph validation is essential when downstream systems depend on semantic data quality.", "Shapes can encode cardinality, datatype, class, and path constraints for RDF data.", "Constraint standards are natural guardrails for LLM-generated graph updates." ], "ontology_relevance": "Critical validation layer for ontology-backed graph construction, enrichment, and maintenance.", "ai_relevance": "Useful guardrail for checking LLM-generated triples, policies, and ontology edits.", "palantir_relevance": "Standards comparator for validation of ontology actions, object edits, and generated semantic data.", "quality_signal": "official_docs", "retrieval_tags": [ "constraints", "llm-guardrails", "shacl-1-2", "validation", "w3c" ], "triage_tier": "candidate", "triage_score": 66 }, { "id": "phase2-benchmarking-agents-2025", "title": "Benchmarking Agentic AI Systems: A Survey", "authors_or_org": "Survey authors", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2503.16416", "doi_or_identifier": "arxiv:2503.16416", "venue_or_site": "arXiv", "abstract_or_summary": "Survey of agentic AI benchmarking, covering task environments, tool use, planning, memory, safety, evaluation metrics, and reproducibility concerns.", "key_claims": [ "Agent benchmarks need to evaluate planning, tool use, memory, safety, reliability, and interaction with environments.", "Tool-call traces and environment state changes are central evidence for agent evaluation.", "Ontology-backed tools create a measurable action space for agents, but require permissioning and audit." ], "ontology_relevance": "Connects ontology-defined action spaces with agent benchmarking and evaluation.", "ai_relevance": "Useful overview for evaluating AI agents beyond static Q&A tasks.", "palantir_relevance": "Relevant to AIP agents, Ontology MCP, AIP Evals, and operational action traces.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "agent-memory", "agents", "benchmark", "evaluation", "governance", "mcp", "tool-use" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase3-dama-dmbok-2017", "title": "DAMA-DMBOK: Data Management Body of Knowledge, Second Edition", "authors_or_org": "DAMA International", "year": 2017, "source_type": "book", "bucket": "technical", "url": "https://technicspub.com/dmbok", "doi_or_identifier": "isbn 9781634622349", "venue_or_site": "Technics Publications", "abstract_or_summary": "Practitioner reference for data management disciplines including governance, architecture, modeling, metadata, quality, master data, and ethics.", "key_claims": [ "Data governance requires roles, responsibilities, policies, standards, and decision rights.", "Metadata, data quality, master data, and architecture are interdependent management disciplines.", "Sustainable data programs require operating models, not just technical platforms." ], "ontology_relevance": "Gives practitioner vocabulary for connecting ontology to data governance, stewardship, metadata, and quality management.", "ai_relevance": "AI governance depends on the maturity of underlying data governance practices.", "palantir_relevance": "Useful comparator for whether enterprise ontology programs cover stewardship, ownership, quality, and lifecycle controls.", "quality_signal": "primary_source", "retrieval_tags": [ "data-governance", "data-quality", "master-data", "metadata", "phase3", "practitioner-framework" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "p2-llm-ont-023", "title": "Graph Retrieval-Augmented Generation: A Survey", "authors_or_org": "Boci Peng; Yun Zhu; Yongchao Liu; Xiaohe Bo; Haizhou Shi; Chuntao Hong; Yan Zhang; Siliang Tang", "year": 2024, "source_type": "survey", "bucket": "academic", "url": "https://arxiv.org/abs/2408.08921", "doi_or_identifier": "arxiv:2408.08921", "venue_or_site": "arXiv", "abstract_or_summary": "Surveys GraphRAG workflows, including graph-based indexing, graph-guided retrieval, graph-enhanced generation, training methods, applications, evaluation, and industrial use cases.", "key_claims": [ "GraphRAG adds relational structure to RAG through graph indexes and graph-aware retrieval.", "Evaluation remains unsettled because graph quality, retrieval quality, and answer quality interact.", "GraphRAG industrial use cases require entity resolution, provenance, and domain modeling." ], "ontology_relevance": "Useful map of graph-augmented retrieval methods adjacent to ontology-backed retrieval.", "ai_relevance": "Shows the emerging retrieval architecture where LLMs use entities, relations, paths, and communities as context.", "palantir_relevance": "Relevant to comparing Palantir-style governed ontology with generated GraphRAG graphs.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "graphrag", "knowledge-graph", "rag", "retrieval", "survey" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "industry-commercial-microsoft-graphrag", "title": "GraphRAG: graph-based retrieval augmented generation", "authors_or_org": "Microsoft Research / Microsoft", "year": 2024, "source_type": "technical_article", "bucket": "technical", "url": "https://github.com/microsoft/graphrag", "doi_or_identifier": "github:microsoft/graphrag", "venue_or_site": "Microsoft GitHub / Microsoft Research", "abstract_or_summary": "Microsoft GraphRAG material describes constructing a knowledge graph from text, detecting communities, summarizing graph structure, and using local/global graph retrieval to improve question answering over private corpora.", "key_claims": [ "Graph structure can support global questions that are difficult for naive chunk retrieval.", "Entity and community summaries provide retrieval units above raw text chunks.", "LLM-generated graphs require careful extraction, provenance, and evaluation controls.", "GraphRAG operationalizes graph extraction, community detection, and summary retrieval as a reusable pipeline.", "Local search and global search encode different retrieval needs over the same generated graph.", "A production GraphRAG workflow needs configuration, prompt control, caching, indexing, and evaluation assets." ], "ontology_relevance": "Shows a lightweight, generated KG approach that differs from curated enterprise ontology but can complement it.", "ai_relevance": "Central source for GraphRAG architectures and graph-based LLM retrieval.", "palantir_relevance": "Useful contrast to Palantir's curated ontology: GraphRAG can generate graph context from documents, but may lack governed object/action semantics.", "quality_signal": "primary_source", "retrieval_tags": [ "communities", "entity-extraction", "global-search", "graphrag", "knowledge-graph", "llm", "local-search", "microsoft", "open-source", "retrieval" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase3-llm-ontmem-023", "title": "LLM-empowered Knowledge Graph Construction: A Survey", "authors_or_org": "LLM-empowered KGC survey authors", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2510.20345", "doi_or_identifier": "arxiv:2510.20345", "venue_or_site": "arXiv", "abstract_or_summary": "Surveys the use of LLMs in knowledge graph construction, including entity and relation extraction, ontology/schema use, alignment, validation, and end-to-end graph building pipelines.", "key_claims": [ "LLM-empowered KGC requires attention to both extraction quality and downstream graph usability.", "Ontology and schema guidance can improve consistency, but hallucination and entity resolution remain core risks.", "Evaluation of KGC pipelines should measure graph-level quality, not just extraction micro-metrics." ], "ontology_relevance": "Synthesis source for how ontology constraints and schemas enter LLM-based KG construction.", "ai_relevance": "Maps LLM roles in automated graph construction and maintenance.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "graph-quality", "knowledge-graph-construction", "llm", "schema-guidance", "survey" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase3-llm-ontmem-012", "title": "LLM-Matcher: A Name-Based Schema Matching Tool using Large Language Models", "authors_or_org": "Marcel Parciak; Brecht Vandevoort; Frank Neven; Liesbet M. Peeters; Stijn Vansummeren", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1145/3722212.3725112", "doi_or_identifier": "10.1145/3722212.3725112", "venue_or_site": "SIGMOD-Companion 2025", "abstract_or_summary": "Presents LLM-Matcher, an interactive name-based schema matching system designed for restricted environments where instance-based matching is not possible, such as healthcare data contexts.", "key_claims": [ "Name and description based matching remains valuable when sensitive domains cannot expose instance data.", "Interactive LLM-assisted matching can support human review instead of fully automated correspondence acceptance.", "Restricted data environments make metadata quality and semantic descriptions central to matching performance." ], "ontology_relevance": "Useful for ontology and schema alignment where privacy prevents using raw records.", "ai_relevance": "Shows LLM matching as a human-facing integration tool under data access constraints.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "healthcare-data", "human-in-the-loop", "llm-matcher", "metadata", "schema-matching" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "industry-commercial-ontotext-case-studies", "title": "Ontotext knowledge graph case studies", "authors_or_org": "Ontotext", "year": 2026, "source_type": "case_study", "bucket": "commercial", "url": "https://www.ontotext.com/customers", "doi_or_identifier": null, "venue_or_site": "Ontotext Customers", "abstract_or_summary": "Ontotext customer materials describe RDF knowledge graph applications in publishing, healthcare, media, life sciences, and enterprise content integration.", "key_claims": [ "Content-rich domains benefit from combining controlled vocabularies, extracted entities, metadata, and provenance.", "Semantic search can improve discovery beyond exact keyword matching.", "RDF graphs support evidence and publication management where source provenance matters." ], "ontology_relevance": "Concrete cases for content and evidence knowledge graphs using semantic web standards.", "ai_relevance": "Provides structured retrieval context for AI over publications, evidence, and enterprise content.", "palantir_relevance": "Shows a standards-based route to domain knowledge integration outside Palantir.", "quality_signal": "primary_source", "retrieval_tags": [ "case-studies", "healthcare", "ontotext", "publishing", "rdf", "semantic-search" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase16-etsi-saref-smart-applications-reference-ontology", "title": "SAREF: Smart Applications REFerence Ontology", "authors_or_org": "ETSI SmartM2M Technical Committee", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://saref.etsi.org", "doi_or_identifier": "etsi ts 103 264", "venue_or_site": "ETSI SAREF Portal", "abstract_or_summary": "SAREF is ETSI's Smart Applications REFerence ontology for IoT, energy, buildings, industry, and smart-city domains. It provides a modular ontology pattern for devices, functions, services, measurements, properties, and domain extensions.", "key_claims": [ "SAREF provides a reusable ontology for smart applications and IoT interoperability.", "The ontology models devices, functions, services, measurements, properties, and domain extensions.", "Operational AI over IoT environments needs domain semantics for sensors, actuators, and measurements." ], "ontology_relevance": "Industrial/IoT ontology standard connecting semantic interoperability to physical devices and measurements.", "ai_relevance": "IoT and smart-building AI systems need shared device and measurement semantics so agents can reason over sensors, actuators, services, and domain-specific extensions.", "palantir_relevance": "Useful non-Palantir comparator for ontology-based operational digital twins and IoT integration.", "quality_signal": "official_standard", "retrieval_tags": [ "digital-twin", "industrial-semantic-standards", "iot", "measurements", "ontology-standard", "phase16", "saref", "semantic-interoperability", "smart-applications" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase4-std-w3c-owl-time-2022", "title": "Time Ontology in OWL", "authors_or_org": "Simon Cox, Chris Little, W3C Spatial Data on the Web Working Group", "year": 2022, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/owl-time", "doi_or_identifier": null, "venue_or_site": "W3C Candidate Recommendation Draft", "abstract_or_summary": "Defines OWL/RDF terms for temporal entities, instants, intervals, durations, temporal reference systems, and temporal relations.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "auditability", "knowledge-graph", "owl", "provenance", "temporal-ontology", "time", "w3c" ], "triage_tier": "candidate", "triage_score": 64 }, { "id": "phase4-palantir-ai-automation-ethics-war-2024", "title": "AI, Automation, and the Ethics of Modern Warfare", "authors_or_org": "Palantir Technologies", "year": 2024, "source_type": "blog", "bucket": "palantir", "url": "https://blog.palantir.com/ai-automation-and-the-ethics-of-modern-warfare-df1f0b212397", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir blog discussing AI, automation, and warfare ethics, relevant to governance claims for defense AI systems.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "widely_cited", "retrieval_tags": [ "automation", "defense-ai", "ethics", "governance", "palantir" ], "triage_tier": "candidate", "triage_score": 62 }, { "id": "phase14-sssom-ontological-mappings-2022", "title": "A Simple Standard for Sharing Ontological Mappings (SSSOM)", "authors_or_org": "Nicolas Matentzoglu; James P. Balhoff; Susan M. Bello; Chris Bizon; Matthew Brush; Tiffany J. Callahan; Christopher G. Chute; et al.; Christopher J. Mungall", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/database/baac035", "doi_or_identifier": "10.1093/database/baac035", "venue_or_site": "Database", "abstract_or_summary": "The SSSOM paper defines a simple standard for sharing ontological mappings. It introduces a machine-readable data model for mapping sets and individual mappings, including required slots such as subject_id, object_id, predicate_id, and match_type, with additional provenance and mapping metadata.", "key_claims": [ "Ontology mappings require machine-readable metadata that makes mapping semantics, provenance, and uncertainty explicit.", "SSSOM models mappings as pairwise assertions grouped into mapping sets with standard slots.", "FAIR exchange of mappings is a prerequisite for reliable ontology integration, reconciliation, and cross-resource data reuse." ], "ontology_relevance": "Strengthens the mapping/alignment layer of the KB: ontologies rarely exist alone, and AI pipelines need traceable mappings across vocabularies and schemas.", "ai_relevance": "Ontology-aware AI systems need transparent, exchangeable mappings between schemas, terminologies, and knowledge graphs; SSSOM gives a concrete metadata contract for those alignments.", "palantir_relevance": "Provides an open standard comparator for enterprise ontology integration and schema alignment across heterogeneous data sources.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "alignment", "fair-data", "mapping-metadata", "ontology-mapping", "phase14", "schema-alignment", "semantic-interoperability", "sssom" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-timmermans-epstein-2010-sociology-standards", "title": "A World of Standards but Not a Standard World: Toward a Sociology of Standards and Standardization", "authors_or_org": "Stefan Timmermans and Steven Epstein", "year": 2010, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1146/annurev.soc.012809.102629", "doi_or_identifier": "10.1146/annurev.soc.012809.102629", "venue_or_site": "Annual Review of Sociology", "abstract_or_summary": "Review article synthesizing sociological research on standards and standardization across professions, markets, medicine, and governance.", "key_claims": [ "Standards coordinate action but also create tensions with local practice and discretion.", "Standardization produces comparability, accountability, and control while reducing some forms of flexibility.", "Standards should be studied through their implementation and effects, not just their formal design." ], "ontology_relevance": "Frames ontology governance as standardization with both coordination benefits and local-practice costs.", "ai_relevance": "AI governance depends on standardized inputs, labels, evaluation criteria, and workflows that may not fit all contexts.", "palantir_relevance": "Useful for discussing how enterprise ontology standardizes operational categories across heterogeneous units.", "quality_signal": "widely_cited", "retrieval_tags": [ "coordination", "local-practice", "phase3", "sociology", "standardization", "standards" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "pal-blog-nvidia-ontology-2025", "title": "AI Infrastructure and Ontology: Palantir and NVIDIA", "authors_or_org": "Palantir Technologies", "year": 2025, "source_type": "commercial_article", "bucket": "palantir", "url": "https://blog.palantir.com/ai-infrastructure-and-ontology-78b86f173ea6", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Vendor-authored partnership article describing NVIDIA models, embeddings, and optimization tools becoming available through Palantir AIP and Foundry/Ontology workflows, with examples in supply chain and healthcare.", "key_claims": [ "NVIDIA models and tools are made available through AIP and Palantir infrastructure.", "NVIDIA NeMo Retriever can support RAG and OAG workflows through Foundry/Ontology tooling.", "The partnership is framed as combining chips, models, and ontology for operational AI." ], "ontology_relevance": "Shows Palantir positioning ontology as part of an AI infrastructure stack.", "ai_relevance": "Connects external model providers, embeddings, optimization, and agentic workflows to AIP/Ontology.", "palantir_relevance": "Primary Palantir commercial narrative for NVIDIA partnership.", "quality_signal": "marketing", "retrieval_tags": [ "aip", "nvidia", "oag", "ontology", "operational-ai", "palantir", "partnership" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase4-palantir-dod-contracts-may-21-2025", "title": "Contracts For May. 21, 2025", "authors_or_org": "U.S. Department of Defense", "year": 2025, "source_type": "government_record", "bucket": "palantir", "url": "https://www.war.gov/News/Contracts/Contract/Article/4194643", "doi_or_identifier": null, "venue_or_site": "U.S. Department of Defense Contracts", "abstract_or_summary": "Official Department of Defense contract notice including Palantir-related award information, useful as public procurement evidence for defense work.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "contract", "defense", "dod", "palantir", "procurement" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-boyd-crawford-2012-critical-big-data", "title": "Critical Questions for Big Data", "authors_or_org": "danah boyd and Kate Crawford", "year": 2012, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1080/1369118X.2012.678878", "doi_or_identifier": "10.1080/1369118x.2012.678878", "venue_or_site": "Information, Communication & Society", "abstract_or_summary": "Widely cited critique asking what big data changes about knowledge, power, ethics, access, and interpretation.", "key_claims": [ "Big data should not be mistaken for objective, complete, or theory-free knowledge.", "Data access and analytic capacity create new power asymmetries.", "Interpretation, context, and ethics remain central even with large-scale data." ], "ontology_relevance": "Counters naive claims that more structured data or bigger knowledge graphs automatically produce better knowledge.", "ai_relevance": "Relevant to LLM and GraphRAG systems that risk presenting incomplete data infrastructure as comprehensive ground truth.", "palantir_relevance": "Useful critique for enterprise ontology and public-sector analytics claims around visibility and objectivity.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-critique", "big-data", "critical-data-studies", "objectivity", "phase3", "power" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-khatri-brown-2010-data-governance", "title": "Designing Data Governance", "authors_or_org": "Vijay Khatri and Carol V. Brown", "year": 2010, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/1629175.1629210", "doi_or_identifier": "10.1145/1629175.1629210", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "Concise practitioner-academic article proposing decision domains for data governance, including principles, quality, metadata, access, and lifecycle.", "key_claims": [ "Data governance is about assigning decision rights and accountability for data assets.", "Key governance domains include data principles, quality, metadata, access, and lifecycle.", "Governance structures should match organizational context and data use." ], "ontology_relevance": "Supports treating ontology governance as decision-rights design around categories, metadata, quality, and access.", "ai_relevance": "AI systems require clear accountability for data definitions, access permissions, and quality thresholds.", "palantir_relevance": "Useful for assessing who controls object types, action types, and semantic changes in enterprise ontology platforms.", "quality_signal": "widely_cited", "retrieval_tags": [ "accountability", "data-governance", "data-quality", "decision-rights", "metadata", "phase3" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase4-std-eu-hleg-trustworthy-ai-guidelines-2019", "title": "Ethics Guidelines for Trustworthy AI", "authors_or_org": "High-Level Expert Group on Artificial Intelligence, European Commission", "year": 2019, "source_type": "policy_guidance", "bucket": "technical", "url": "https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai", "doi_or_identifier": null, "venue_or_site": "European Commission High-Level Expert Group Report", "abstract_or_summary": "European Commission expert-group guidance defining trustworthy AI through lawful, ethical, and robust AI, with requirements such as human agency, technical robustness, privacy, transparency, fairness, societal wellbeing, and accountability.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "accountability", "ethics", "eu", "governance", "human-oversight", "transparency", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase14-fair-data-point-metadata-publication-2023", "title": "FAIR Data Point: A FAIR-Oriented Approach for Metadata Publication", "authors_or_org": "Luiz Olavo Bonino da Silva Santos; Kees Burger; Rajaram Kaliyaperumal; Mark D. Wilkinson", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1162/dint_a_00160", "doi_or_identifier": "10.1162/dint_a_00160", "venue_or_site": "Data Intelligence", "abstract_or_summary": "The FAIR Data Point paper presents a software architecture for publishing semantically rich and machine-actionable metadata according to FAIR principles. It describes core components, metadata provision, adherence criteria, registration, indexing, and search for FAIR metadata content.", "key_claims": [ "FAIR metadata publication requires semantically rich, machine-actionable metadata services.", "FAIR Data Point defines architecture and metadata provision patterns for discovering and searching metadata about digital objects.", "The approach connects linked data, FAIR principles, metadata schemas, and service-level discovery." ], "ontology_relevance": "Strengthens the machine-actionable metadata route: ontologies and linked data need service interfaces and metadata publication mechanisms, not only local schemas.", "ai_relevance": "Agentic AI and scientific AI need discoverable, semantically rich metadata services for datasets, tools, algorithms, and digital objects; FAIR Data Point is a concrete architecture for this layer.", "palantir_relevance": "Useful comparator for catalog/ontology metadata services and governed discovery across enterprise data products.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "dcat", "fair-data", "fair-data-point", "linked-data", "machine-actionable-metadata", "metadata-service", "phase14", "semantic-interoperability" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-selbst-etal-2019-fairness-abstraction", "title": "Fairness and Abstraction in Sociotechnical Systems", "authors_or_org": "Andrew D. Selbst, Danah Boyd, Sorelle A. Friedler, Suresh Venkatasubramanian, and Janet Vertesi", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3287560.3287598", "doi_or_identifier": "10.1145/3287560.3287598", "venue_or_site": "ACM Conference on Fairness, Accountability, and Transparency", "abstract_or_summary": "Highly cited critique arguing that ML fairness work often fails by abstracting away the social systems in which algorithms operate.", "key_claims": [ "Fairness failures often arise when technical abstractions ignore social context.", "Systems are embedded in institutions, feedback loops, and power relations.", "Technical metrics cannot substitute for socio-technical analysis of deployment contexts." ], "ontology_relevance": "Warns against treating ontology quality as only logical consistency or schema cleanliness.", "ai_relevance": "Strong source for arguing that ontology-grounded AI must be evaluated in institutional context.", "palantir_relevance": "Relevant to public-sector deployments where object models and AI workflows interact with real institutions and affected groups.", "quality_signal": "widely_cited", "retrieval_tags": [ "abstraction-trap", "ai-governance", "critique", "fairness", "phase3", "sociotechnical-systems" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase6-nhs-fdp-contracts-finder-2024", "title": "Federated Data Platform and Associated Services", "authors_or_org": "NHS England / UK Contracts Finder", "year": 2024, "source_type": "webpage", "bucket": "palantir", "url": "https://www.contractsfinder.service.gov.uk/Notice/0f8a65b5-23a2-4294-abb1-a7fd8efb3ad0", "doi_or_identifier": "cf-2082000d0o000000rwimuaa", "venue_or_site": "Contracts Finder", "abstract_or_summary": "Official award notice for the NHS Federated Data Platform and Associated Services contract awarded to Palantir Technologies UK Ltd, with contract documents attached and a stated base value of GBP 182.2m, forecastable to GBP 330m.", "key_claims": [ "The award notice describes a cloud-based SaaS solution for NHS bodies.", "Palantir Technologies UK Ltd was awarded the FDP contract.", "The notice lists a contract end date of 2027-02-15.", "The notice states a base contract value and forecastable value." ], "ontology_relevance": "Primary procurement evidence for the platform that operationalizes the NHS FDP data model and application layer.", "ai_relevance": "Baseline source for assessing AI-enabled platform claims against actual contracted service scope.", "palantir_relevance": "Direct Palantir FDP contract award record.", "quality_signal": "primary_source", "retrieval_tags": [ "contract-award", "contracts-finder", "fdp", "nhs", "palantir", "phase6", "procurement", "public-sector" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase6-sui-2025-fidelis-kgqa", "title": "FiDeLiS: Faithful Reasoning in Large Language Models for Knowledge Graph Question Answering", "authors_or_org": "Yuan Sui; Yufei He; Nian Liu; Xiaoxin He; Kun Wang; Bryan Hooi", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://aclanthology.org/2025.findings-acl.436", "doi_or_identifier": "10.18653/v1/2025.findings-acl.436", "venue_or_site": "Findings of ACL 2025", "abstract_or_summary": "Grounds LLM KGQA in verifiable KG reasoning paths using high-recall retrieval, candidate path construction, and deductive validation.", "key_claims": [ "Verifiable reasoning paths reduce hallucination in KGQA.", "Training-free path retrieval and beam search can improve accuracy and efficiency.", "LLMs can score reasoning paths as deductive reasoning tasks." ], "ontology_relevance": "Moderate-to-strong: uses KG structure and relation paths, though less explicit about formal ontologies.", "ai_relevance": "Important for LLM reasoning faithfulness over structured knowledge.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "acl-2025", "faithful-reasoning", "kgqa", "knowledge-graph", "phase6", "reasoning-paths", "structured-grounding" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "guarino_1998_fois", "title": "Formal Ontology and Information Systems", "authors_or_org": "Nicola Guarino", "year": 1998, "source_type": "paper", "bucket": "academic", "url": "https://ontolog.cim3.net/file/resource/historic-archives/FOIS-community/Guarino98_Formal-Ontology-and-Information-Systems_FOIS-1998_NicolaGuarino_19980606to08.pdf", "doi_or_identifier": "fois 1998 proceedings paper", "venue_or_site": "Formal Ontology in Information Systems (FOIS 1998)", "abstract_or_summary": "Foundational FOIS paper arguing that formal ontology clarifies the intended meaning of information-system vocabularies and supports better system design and integration.", "key_claims": [ "Ontology is a formal account of intended meaning for a vocabulary.", "Information systems benefit when conceptual assumptions are explicit.", "Formal ontology helps separate conceptual analysis from implementation artifacts." ], "ontology_relevance": "One of the central papers establishing formal ontology as a discipline for information systems.", "ai_relevance": "Important for AI systems that need explicit semantic grounding rather than only database schemas or labels.", "palantir_relevance": "Directly relevant to interpreting enterprise ontology as an operational semantic layer over heterogeneous systems.", "quality_signal": "widely_cited", "retrieval_tags": [ "fois", "formal-ontology", "information-systems", "intended-meaning" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "hitzler_krotzsch_rudolph_2009_semantic_web_foundations", "title": "Foundations of Semantic Web Technologies", "authors_or_org": "Pascal Hitzler, Markus Krotzsch, and Sebastian Rudolph", "year": 2009, "source_type": "book", "bucket": "books", "url": "https://doi.org/10.1201/9781420090512", "doi_or_identifier": "10.1201/9781420090512", "venue_or_site": "CRC Press", "abstract_or_summary": "Formal introduction to RDF, RDFS, OWL, rules, querying, and reasoning in Semantic Web systems.", "key_claims": [ "Semantic Web technologies combine graph data with formal semantics and inference.", "RDF, RDFS, OWL, and rules occupy different expressivity and reasoning positions.", "Formal foundations are necessary for interoperable machine-processable meaning." ], "ontology_relevance": "Connects standards with formal reasoning foundations for ontology-based systems.", "ai_relevance": "Useful bridge from symbolic ontology to modern knowledge graph and reasoning architectures.", "palantir_relevance": "Relevant to evaluating how much formal Semantic Web machinery an operational ontology needs.", "quality_signal": "widely_cited", "retrieval_tags": [ "owl", "rdf", "reasoning", "rules", "semantic-web" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "p2-llm-ont-031", "title": "From Instructions to ODRL Usage Policies: An Ontology Guided Approach", "authors_or_org": "Daham M. Mustafa; Abhishek Nadgeri; Diego Collarana; Benedikt T. Arnold; Christoph Quix; Christoph Lange; Stefan Decker", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2506.03301", "doi_or_identifier": "arxiv:2506.03301", "venue_or_site": "LLM+KG Workshop at VLDB 2024 / arXiv", "abstract_or_summary": "Uses LLMs and ODRL ontology documentation to generate machine-readable usage policy knowledge graphs from natural language instructions in dataspace scenarios.", "key_claims": [ "Ontology documentation can guide LLMs to produce more valid policy graphs.", "Natural language policy instructions need semantic constraints before operational use.", "Policy generation quality improves when prompts are grounded in curated ontology terms." ], "ontology_relevance": "Demonstrates ontology-guided LLM generation for governance and usage-policy semantics.", "ai_relevance": "Shows structured policy generation as an LLM plus ontology use case.", "palantir_relevance": "Relevant to controlled AI actions, data entitlements, and policy-aware semantic layers.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "dataspaces", "llm", "odrl", "ontology-guided-generation", "usage-policy" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "p2-llm-ont-030", "title": "Integrating Large Language Models and Knowledge Graphs for Extraction and Validation of Textual Test Data", "authors_or_org": "Antonio De Santis; Marco Balduini; Federico De Santis; Andrea Proia; Arsenio Leo; Marco Brambilla; Emanuele Della Valle", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2408.01700", "doi_or_identifier": "10.1007/978-3-031-77847-6_17", "venue_or_site": "ISWC 2024 In-Use Track / arXiv", "abstract_or_summary": "Industrial case study using knowledge graphs and LLMs to extract and validate textual aerospace test data, extending the Semantic Sensor Network ontology and using virtual knowledge graph access.", "key_claims": [ "LLM extraction becomes more useful when validated against a KG and domain ontology.", "Virtual knowledge graphs can integrate structured and unstructured test data without full physical consolidation.", "Industrial LLM use cases need cost-benefit analysis alongside accuracy." ], "ontology_relevance": "Concrete in-use evidence for ontology-guided extraction and validation in engineering documents.", "ai_relevance": "Shows LLM plus KG workflow for high-complexity, low-volume industrial data.", "palantir_relevance": "Very relevant to operational ontology over manufacturing, testing, and validation workflows.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "industrial-ai", "iswc-2024", "knowledge-graph", "llm", "ontology-guided-extraction" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "ont-ai-025", "title": "Integrating Ontologies and Large Language Models to Implement Retrieval-Augmented Generation", "authors_or_org": "Fernando Gutiérrez; et al.", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://content.iospress.com/articles/applied-ontology/ao230395", "doi_or_identifier": "10.3233/ao-230395", "venue_or_site": "Applied Ontology", "abstract_or_summary": "Explores how formal ontologies can be integrated with LLMs to implement RAG workflows with more structured retrieval and grounding.", "key_claims": [ "Ontologies can structure and constrain RAG context.", "LLMs benefit from explicit domain semantics when retrieving and generating answers.", "Ontology integration introduces engineering overhead and requires careful evaluation." ], "ontology_relevance": "Direct source for ontology-enhanced RAG.", "ai_relevance": "Shows a hybrid LLM plus ontology retrieval architecture.", "palantir_relevance": "Highly relevant to enterprise ontology-backed assistants and workflows.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "applied-ontology", "llm", "ontology-rag", "structured-retrieval" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "p2-llm-ont-021", "title": "Large Language Models and Knowledge Graphs: Opportunities and Challenges", "authors_or_org": "Jeff Z. Pan; Simon Razniewski; Jan-Christoph Kalo; Sneha Singhania; Jiaoyan Chen; Stefan Dietze; Hajira Jabeen; Janna Omeliyanenko; Wen Zhang; Matteo Lissandrini; Russa Biswas; Gerard de Melo; Angela Bonifati; Edlira Vakaj; Mauro Dragoni; Damien Graux", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://drops.dagstuhl.de/entities/document/10.4230/TGDK.1.1.2", "doi_or_identifier": "10.4230/tgdk.1.1.2", "venue_or_site": "Transactions on Graph Data and Knowledge", "abstract_or_summary": "Position paper on opportunities and challenges at the intersection of LLMs, knowledge graphs, ontologies, retrieval-augmented language models, and hybrid explicit plus parametric knowledge.", "key_claims": [ "The LLM era shifts knowledge representation toward hybrid explicit and parametric knowledge.", "KGs and ontologies remain valuable for factual grounding, structure, and symbolic reasoning.", "LLMs create new opportunities for KG construction, enrichment, and interaction." ], "ontology_relevance": "High-quality conceptual framing for why ontologies remain relevant after frontier LLMs.", "ai_relevance": "Connects LLM strengths and weaknesses to explicit graph knowledge infrastructure.", "palantir_relevance": "Useful neutral literature support for ontology as the governance and grounding layer beneath operational AI.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "hybrid-knowledge", "knowledge-graph", "llm", "ontology", "tgdk" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase14-linkml-open-data-modeling-framework-2026", "title": "LinkML: an open data modeling framework", "authors_or_org": "Sierra A. T. Moxon; Harold Solbrig; Nomi L. Harris; Patrick Kalita; Mark A. Miller; Sujay Patil; Kevin Schaper; Chris Bizon; J. Harry Caufield; et al.; LinkML Community Contributors", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/gigascience/giaf152", "doi_or_identifier": "10.1093/gigascience/giaf152", "venue_or_site": "GigaScience", "abstract_or_summary": "LinkML is a linked data modeling framework for authoring, validating, and sharing data schemas. The GigaScience technical note describes LinkML as a bridge between practical schema engineering and semantic-web/FAIR data requirements, supporting YAML-authored models, validation, generators, inheritance, and machine-readable semantics.", "key_claims": [ "LinkML provides an approachable modeling framework for authoring, validating, and sharing structured data schemas.", "The framework connects software-engineering schema needs with linked-data and ontology-ready semantics.", "Typed schemas and validation help make scientific and enterprise data more reusable, interoperable, and less ambiguous for downstream computation." ], "ontology_relevance": "Directly supports the article's claim that ontology-backed AI requires schema languages that operationalize classes, slots, identifiers, constraints, mappings, and validation in everyday data work.", "ai_relevance": "Strong evidence that AI-ready datasets and agent workflows need typed, validated, machine-readable schemas rather than informal documentation alone.", "palantir_relevance": "Useful comparator for Palantir Ontology object/action modeling: LinkML shows an open, community-oriented way to encode typed domain models and validation rules.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "data-modeling", "fair-data", "linkml", "machine-actionable-schema", "ontology-engineering", "phase14", "schema-language", "semantic-interoperability", "validation" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase16-miro-guidelines-minimum-information-reporting-ontology-2017", "title": "MIRO: Guidelines for Minimum Information for the Reporting of an Ontology", "authors_or_org": "Robert Stevens; et al.", "year": 2017, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1186/s13326-017-0172-7", "doi_or_identifier": "10.1186/s13326-017-0172-7", "venue_or_site": "Journal of Biomedical Semantics", "abstract_or_summary": "The MIRO guidelines specify minimum information that should be reported with an ontology, including motivation, scope, knowledge acquisition, ontology content, evaluation, reuse, maintenance, and availability. They make ontology reporting and peer review more reproducible.", "key_claims": [ "Ontology publications should report enough information for reuse, evaluation, and comparison.", "Scope, intended use, knowledge acquisition, reuse, evaluation, and maintenance are part of responsible ontology reporting.", "Transparent ontology reporting supports scientific reuse and downstream computational trust." ], "ontology_relevance": "High-value reporting guideline for ontology evaluation, publication, and governance.", "ai_relevance": "AI systems that publish or depend on ontologies need transparent reporting of scope, evaluation, reuse, and maintenance so generated or curated models can be assessed by other agents and humans.", "palantir_relevance": "Provides a neutral checklist for evaluating how much operational ontology documentation is publicly inspectable.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "documentation", "miro", "ontology-evaluation", "ontology-governance", "ontology-reporting", "phase16", "reporting-guidelines", "reproducibility" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase15-arenas-2024-morph-kgc-scalable-kg-materialization", "title": "Morph-KGC: Scalable knowledge graph materialization with mapping partitions", "authors_or_org": "Julián Arenas-Guerrero; David Chaves-Fraga; Jhon Toledo; María S. Pérez; Óscar Corcho", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/SW-223135", "doi_or_identifier": "10.3233/sw-223135", "venue_or_site": "Semantic Web", "abstract_or_summary": "Morph-KGC is a knowledge graph construction engine that materializes RDF knowledge graphs from heterogeneous sources using R2RML/RML mappings. The Semantic Web article introduces mapping partitions and optimizations for scalable KG materialization.", "key_claims": [ "Morph-KGC constructs RDF knowledge graphs from heterogeneous data sources using R2RML and RML.", "Mapping partitions reduce execution time and memory consumption for large knowledge graph materialization tasks.", "Knowledge graph construction needs scalable execution engines in addition to mapping languages." ], "ontology_relevance": "Supports the engineering claim that ontology-backed AI depends on repeatable, scalable graph materialization and mapping execution pipelines.", "ai_relevance": "A concrete system source showing that ontology/KG-backed AI requires scalable, engineered construction pipelines, not just conceptual graph models.", "palantir_relevance": "Useful implementation comparator for enterprise data integration pipelines that feed operational ontologies or semantic layers.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "heterogeneous-data", "knowledge-graph-construction", "mapping-language", "morph-kgc", "phase15", "rdf", "rml", "scalable-kg-construction" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase4-palantir-nhs-fdp-market-engagement-2023", "title": "NHS Federated Data Platform and Associated Services", "authors_or_org": "NHS England", "year": 2023, "source_type": "government_record", "bucket": "palantir", "url": "https://www.england.nhs.uk/wp-content/uploads/2023/05/PRN00389-nhs-federated-data-platform-and-associated-services.pdf", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "NHS England procurement/market-engagement document describing requirements for the Federated Data Platform and associated services.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "nhs", "palantir", "procurement", "public-sector", "requirements" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase4-palantir-nhs-fdp-prospectus-2023", "title": "NHS Federated Data Platform and Associated Services Document 1 Prospectus", "authors_or_org": "NHS England", "year": 2023, "source_type": "government_record", "bucket": "palantir", "url": "https://atamis-1928.my.salesforce-sites.com/servlet/servlet.FileDownload?file=00P8d000009dbn6EAA", "doi_or_identifier": null, "venue_or_site": "NHS England / Atamis procurement portal", "abstract_or_summary": "Procurement prospectus for NHS Federated Data Platform and associated services, giving buyer-side context for platform scope and expected capabilities.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "nhs", "palantir", "procurement", "prospectus" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase7-obo-foundry-2021-operationalizing-open-data-principles", "title": "OBO Foundry in 2021: Operationalizing Open Data Principles to Evaluate Ontologies", "authors_or_org": "OBO Foundry contributors", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/database/baab069", "doi_or_identifier": "10.1093/database/baab069", "venue_or_site": "Database", "abstract_or_summary": "Evaluates OBO Foundry ontologies against open data and ontology governance principles, operationalizing openness, common syntax, orthogonality, versioning, identifiers, definitions, relations, and maintenance practices.", "key_claims": [ "Ontology ecosystems need measurable governance principles.", "Open biomedical ontologies are evaluated across identifiers, definitions, relations, versioning, openness, and maintenance.", "Interoperability depends on governance and lifecycle discipline, not only representation language." ], "ontology_relevance": "Directly supports the article claim that ontology quality is an institutional process and not just graph construction.", "ai_relevance": "Gives AI systems a governance template for interoperable domain ontologies used in biomedical data integration and machine-actionable knowledge.", "palantir_relevance": "Useful comparator for enterprise ontology governance, modular ownership, versioning, and shared relation design.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "governance", "interoperability", "obo-foundry", "ontology-governance", "phase7", "scientific-infrastructure", "versioning" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase6-sharma-2025-og-rag", "title": "OG-RAG: Ontology-Grounded Retrieval-Augmented Generation for Large Language Models", "authors_or_org": "Kartik Sharma; Peeyush Kumar; Yunqing Li", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://aclanthology.org/2025.emnlp-main.1674", "doi_or_identifier": "10.18653/v1/2025.emnlp-main.1674", "venue_or_site": "EMNLP 2025 / ACL Anthology", "abstract_or_summary": "Introduces an ontology-grounded RAG method that builds ontology-informed hypergraph contexts for fact-based LLM answering.", "key_claims": [ "Domain ontologies can improve retrieval context construction.", "The method reports gains in accurate fact recall, response correctness, attribution speed, and fact-based reasoning.", "Ontology-grounded RAG targets rule- and workflow-heavy domains where generic RAG underperforms." ], "ontology_relevance": "Directly uses domain ontologies to constrain and organize retrieval context.", "ai_relevance": "Directly relevant to GraphRAG/KG-RAG and grounding LLM generation in structured domain semantics.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "domain-ontology", "emnlp-2025", "graphrag", "hypergraph-retrieval", "og-rag", "ontology-grounded-rag", "phase6", "rag" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase7-ontology-development-kit-toolkit-2022", "title": "Ontology Development Kit: a toolkit for building, maintaining and standardizing biomedical ontologies", "authors_or_org": "Nicolas Matentzoglu; Nomi L. Harris; James P. Balhoff; et al.", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/database/baac087", "doi_or_identifier": "10.1093/database/baac087", "venue_or_site": "Database", "abstract_or_summary": "Presents the Ontology Development Kit as tooling for building, maintaining, testing, releasing, and standardizing biomedical ontologies using templates, automated workflows, and OBO community practices.", "key_claims": [ "Ontology development requires repeatable workflows for release, testing, imports, and documentation.", "ODK packages community best practices into reusable project templates.", "Automated ontology engineering tooling makes ontology maintenance more scalable and auditable." ], "ontology_relevance": "Practical infrastructure source for ontology lifecycle, release engineering, and quality control.", "ai_relevance": "Shows how ontology engineering can be made reproducible and automation-friendly, which is essential when LLMs assist ontology construction and updates.", "palantir_relevance": "Provides an open-source analogue for enterprise ontology lifecycle governance and automated validation.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "maintenance", "obo", "ontology-development-kit", "ontology-engineering", "phase7", "reproducibility", "tooling" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "p2-llm-ont-005", "title": "Ontology Generation using Large Language Models", "authors_or_org": "Anna Sofia Lippolis; Mohammad Javad Saeedizade; Robin Keskisarkka; Sara Zuppiroli; Miguel Ceriani; Aldo Gangemi; Eva Blomqvist; Andrea Giovanni Nuzzolese", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1007/978-3-031-94575-5_18", "doi_or_identifier": "10.1007/978-3-031-94575-5_18", "venue_or_site": "arXiv", "abstract_or_summary": "Evaluates prompting techniques for generating OWL ontology drafts from user stories and competency questions, including Ontogenia and Memoryless CQbyCQ.", "key_claims": [ "Competency questions and user stories can guide LLMs toward more usable ontology drafts.", "Ontology quality needs multi-dimensional evaluation, including structure and expert judgment.", "LLM-generated ontologies still show variability and modeling mistakes.", "Requirements expressed as user stories and competency questions can guide LLMs toward more useful ontology drafts.", "Ontology output quality must be assessed with structural metrics and expert qualitative review.", "Even strong LLM ontology drafts show variability and modeling errors that require engineering oversight." ], "ontology_relevance": "Highly relevant to using LLMs as ontology authoring assistants under requirements-driven engineering.", "ai_relevance": "Connects prompt design to formal OWL output and expert evaluation.", "palantir_relevance": "Maps well to enterprise workflows where ontology changes should start from business requirements and competency questions.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "competency-questions", "evaluation", "llm", "ontogenia", "ontology-generation", "owl" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase2-graphrag-017", "title": "Ontology-grounded Automatic Knowledge Graph Construction by LLM under Wikidata schema", "authors_or_org": "Xiaohan Feng; Xixin Wu; Helen Meng", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2412.20942", "doi_or_identifier": "arxiv:2412.20942", "venue_or_site": "arXiv", "abstract_or_summary": "Explores using an existing ontology-like schema such as Wikidata to ground LLM-based knowledge graph construction and reduce unconstrained relation generation.", "key_claims": [ "Schema grounding can reduce relation drift in LLM-generated KGs.", "Existing ontology classes and properties provide reusable constraints for extraction.", "Automatic graph construction still needs validation, provenance, and entity resolution." ], "ontology_relevance": "Strongly relevant to the ontology argument because it treats schema as a control surface for LLM extraction.", "ai_relevance": "Useful for LLM-assisted KG construction and schema-constrained extraction.", "palantir_relevance": "Maps to enterprise ontology population where object and link types should constrain extraction.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "kg-construction", "knowledge-graph-construction", "llm", "llm-extraction", "ontology-grounded-extraction", "ontology-grounded-kg", "schema-constraints", "wikidata", "wikidata-schema" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase11-ali-2026-ontology-grounded-kg-clinical-hallucinations", "title": "Ontology-grounded knowledge graphs for mitigating hallucinations in large language models for clinical question answering", "authors_or_org": "Mohamed Ali; Zaki Taha; Mohamed Mabrouk Morsey", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.jbi.2026.104993", "doi_or_identifier": "10.1016/j.jbi.2026.104993", "venue_or_site": "Journal of Biomedical Informatics", "abstract_or_summary": "Reports an ontology-grounded knowledge-graph approach for mitigating hallucinations in clinical question answering with large language models. The DOI metadata identifies the article as a 2026 Journal of Biomedical Informatics paper, making it a high-value peer-reviewed anchor for the claim that ontology/KG grounding is moving into safety-critical biomedical QA.", "key_claims": [ "Ontology-grounded knowledge graphs are being evaluated as a mitigation strategy for hallucinations in clinical LLM question answering.", "Biomedical QA is a high-consequence setting where semantic grounding, provenance, and domain constraints matter more than open-ended language fluency.", "The article's reference network connects ontology-grounded QA to SPARQL generation, KG-guided RAG, COVID-19 knowledge graphs, and dense passage retrieval." ], "ontology_relevance": "Strong current evidence that ontologies and KGs are being used as safety-oriented semantic infrastructure in clinical AI.", "ai_relevance": "Direct evidence for using ontology-grounded knowledge graphs to reduce hallucination risk in clinical LLM question answering.", "palantir_relevance": "Provides an independent health-domain comparator for Palantir-style claims about governed semantic layers; it should be used as academic evidence, not as Palantir validation.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "clinical-qa", "hallucination-mitigation", "knowledge-graph", "llm", "ontology-grounded-rag", "phase11", "rag", "safety-critical-ai" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase5-liu-2025-ontology-guided-reverse-thinking-kgqa", "title": "Ontology-Guided Reverse Thinking Makes Large Language Models Stronger on Knowledge Graph Question Answering", "authors_or_org": "Runxuan Liu; Bei Luo; Jiaqi Li; Baoxin Wang; Ming Liu; Dayong Wu; Shijin Wang; Bing Qin", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://aclanthology.org/2025.acl-long.741", "doi_or_identifier": "10.18653/v1/2025.acl-long.741", "venue_or_site": "ACL 2025", "abstract_or_summary": "ACL 2025 paper proposing Ontology-Guided Reverse Thinking for knowledge graph question answering, using LLM-extracted purpose and condition labels, ontology-based label reasoning paths, and guided knowledge retrieval.", "key_claims": [ "Knowledge graph question answering still struggles with multi-hop reasoning when entity vector matching misses abstract question purposes.", "Ontology-Guided Reverse Thinking constructs reasoning paths from purposes back to conditions.", "The method uses LLMs to extract purpose and condition labels, constructs label reasoning paths based on KG ontology, and uses those paths to guide retrieval.", "Experiments on WebQSP and CWQ report state-of-the-art performance and improved LLM KGQA capability." ], "ontology_relevance": "Strong peer-reviewed source showing ontology structure guiding LLM reasoning and retrieval for multi-hop KGQA.", "ai_relevance": "Useful for arguing that ontology is not only storage but a reasoning-path scaffold for LLM retrieval and question answering.", "palantir_relevance": "Conceptual comparator for ontology-guided retrieval in enterprise AI, though not Palantir-specific.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "acl-2025", "kgqa", "knowledge-graph", "llm", "multi-hop-reasoning", "ontology", "ontology-guided", "phase5", "retrieval" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase14-calvanese-2016-ontop-answering-sparql-relational-databases", "title": "Ontop: Answering SPARQL queries over relational databases", "authors_or_org": "Diego Calvanese; Benjamin Cogrel; Sarah Komla-Ebri; Roman Kontchakov; Davide Lanti; Martin Rezk; Mariano Rodriguez-Muro; Guohui Xiao", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/SW-160217", "doi_or_identifier": "10.3233/sw-160217", "venue_or_site": "Semantic Web", "abstract_or_summary": "The Ontop framework demonstrates ontology-based data access over relational databases, using a conceptual RDF(S)/OWL ontology layer plus mappings to answer SPARQL queries without materializing all source data. Ontop is a concrete virtual knowledge graph system lineage for OBDA.", "key_claims": [ "Ontop implements OBDA using an ontology layer, relational sources, and mapping assertions.", "Users can query databases as virtual RDF graphs through SPARQL and OWL/RDF(S)-based conceptual models.", "Virtual knowledge graph systems decouple user-facing semantics from physical storage schemas." ], "ontology_relevance": "Makes the OBDA/virtual-KG concept concrete and retrievable for agents needing implementation-level references.", "ai_relevance": "Concrete system evidence for AI and analytics architectures that expose governed data through a semantic virtual layer instead of pushing language models directly onto physical tables.", "palantir_relevance": "Useful non-Palantir comparator for ontology-mediated access over databases and enterprise semantic layers.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ontology-based-data-access", "ontop", "phase14", "rdf", "semantic-layer", "sparql", "virtual-knowledge-graph" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase2-poveda-2014-oops", "title": "OOPS! (OntOlogy Pitfall Scanner!): An On-line Tool for Ontology Evaluation", "authors_or_org": "Maria Poveda-Villalon; Asuncion Gomez-Perez; Mari Carmen Suarez-Figueroa", "year": 2014, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/s13740-014-0028-y", "doi_or_identifier": "10.1007/s13740-014-0028-y", "venue_or_site": "International Journal on Semantic Web and Information Systems", "abstract_or_summary": "Enhances the existing OOPS! record as a practical ontology quality-assurance source. OOPS! detects common ontology modeling pitfalls and gives ontology engineers automated warnings that complement human expert review and competency-question evaluation.", "key_claims": [ "Ontology quality can be partially evaluated by detecting recurring modeling pitfalls.", "Automated pitfall scanning complements expert review, reasoner checks, and competency-question tests.", "LLM-accelerated ontology engineering increases the need for scalable quality gates.", "Ontology quality can be partially evaluated by detecting common modeling pitfalls.", "Pitfall scanning complements human expert review and competency-question testing.", "Automated ontology evaluation is essential when LLMs accelerate ontology generation." ], "ontology_relevance": "Important practical source for ontology quality assurance, pitfall detection, and evaluation tooling.", "ai_relevance": "LLM-generated ontologies and graphs need automatic pitfall checks before being used as retrieval, reasoning, or agent-action infrastructure.", "palantir_relevance": "Relevant to quality controls for enterprise ontology schema evolution and AI-assisted object modeling.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "automated-quality-check", "ontology-engineering", "ontology-evaluation", "ontology-governance", "oops", "phase16", "pitfalls", "quality", "validation" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase11-hamed-rocha-2026-biomedical-rag-majority-voting-verification-protocol", "title": "Protocol for evaluating ChatGPT in biomedical association generation and verification using a RAG-enabled, cross-model majority voting workflow", "authors_or_org": "Ahmed Abdeen Hamed; Luis M. Rocha", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.xpro.2026.104533", "doi_or_identifier": "10.1016/j.xpro.2026.104533", "venue_or_site": "STAR Protocols", "abstract_or_summary": "Protocol paper for evaluating ChatGPT in biomedical association generation and verification with a RAG-enabled, cross-model majority-voting workflow. Crossref metadata links it to biomedical knowledge verification work and includes ontology-driven biological graph fact-checking in the reference chain.", "key_claims": [ "Biomedical generative claims require verification workflows rather than direct acceptance of LLM outputs.", "RAG-enabled cross-model majority voting is proposed as a protocol for evaluating biomedical association generation and verification.", "The work connects to ontology-driven biological graph fact-checking and biomedical knowledge-network verification.", "The protocol is useful as evaluation methodology evidence, not as a general guarantee that voting removes hallucination." ], "ontology_relevance": "Provides biomedical verification context adjacent to ontology-driven biological graph fact-checking and RAG-based hallucination mitigation.", "ai_relevance": "Evidence for biomedical LLM verification workflows that combine retrieval, model plurality, and knowledge-network fact checking.", "palantir_relevance": "Indirect comparator for evaluation and verification workflows in high-consequence domains; not Palantir evidence.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "biomedical-verification", "cross-model-voting", "evaluation-protocol", "fact-checking", "hallucination-mitigation", "phase11", "rag", "validation" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "ont-ai-049", "title": "Question Answering over Knowledge Graphs: A Survey", "authors_or_org": "Dennis Diefenbach; Vanessa Lopez; Kamal Singh; Pierre Maret", "year": 2018, "source_type": "paper", "bucket": "academic", "url": "https://content.iospress.com/articles/semantic-web/sw222", "doi_or_identifier": "10.3233/sw-170270", "venue_or_site": "Semantic Web", "abstract_or_summary": "Survey of approaches and benchmarks for answering natural-language questions over knowledge graphs.", "key_claims": [ "KGQA requires entity linking, relation detection, query construction, and answer ranking.", "Natural language interfaces are a major way users access KGs.", "System performance depends on both language understanding and graph quality." ], "ontology_relevance": "Directly relevant to natural-language access to ontologies/KGs.", "ai_relevance": "Important survey for structured retrieval and KG question answering.", "palantir_relevance": "Relevant to asking natural-language questions over an enterprise ontology.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "kgqa", "knowledge-graph", "natural-language-interface", "question-answering" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase2-ragtruth-2024", "title": "RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models", "authors_or_org": "Cheng Niu; Yuanhao Wu; Juno Zhu; Siliang Xu; Kashun Shum; Randy Zhong; Juntong Song; Tong Zhang", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2401.00396", "doi_or_identifier": "arxiv:2401.00396", "venue_or_site": "ACL 2024 / arXiv", "abstract_or_summary": "Presents a corpus of approximately 18,000 RAG responses with detailed hallucination annotations across tasks and domains.", "key_claims": [ "RAG reduces but does not eliminate unsupported or contradictory generated claims.", "Word-level hallucination annotation helps evaluate grounding failures more precisely.", "Trustworthy ontology+AI systems need evidence-grounding checks even when retrieval is present." ], "ontology_relevance": "Supports claim-level evaluation for ontology-grounded generation and graph RAG outputs.", "ai_relevance": "Important hallucination dataset for trustworthy RAG research.", "palantir_relevance": "Relevant to evaluating AI outputs that rely on Palantir ontology context or retrieved evidence.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "evaluation", "evidence", "faithfulness", "hallucination", "rag", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "industry-commercial-rdfox-reasoning", "title": "RDFox documentation and product materials", "authors_or_org": "Oxford Semantic Technologies", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://www.oxfordsemantic.tech/product", "doi_or_identifier": null, "venue_or_site": "Oxford Semantic Technologies", "abstract_or_summary": "RDFox materials describe a high-performance RDF graph and reasoning engine supporting rule-based inference over knowledge graphs.", "key_claims": [ "High-performance reasoning can make semantic rules practical for operational graph workloads.", "Datalog-style rules and RDF semantics can derive new facts from asserted graph data.", "Reasoning performance is a differentiator for ontology-heavy enterprise applications." ], "ontology_relevance": "Useful for ontology reasoning, inference, and rule execution patterns.", "ai_relevance": "Reasoned facts can improve AI retrieval and decision-support context.", "palantir_relevance": "Alternative to proprietary derived object logic in ontology platforms.", "quality_signal": "official_docs", "retrieval_tags": [ "datalog", "inference", "rdf", "rdfox", "reasoning", "rules" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase7-robot-automating-ontology-workflows-2019", "title": "ROBOT: A Tool for Automating Ontology Workflows", "authors_or_org": "Robert C. Jackson; James P. Balhoff; Emily Douglass; Nomi L. Harris; Christopher J. Mungall; James A. Overton", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1186/s12859-019-3002-3", "doi_or_identifier": "10.1186/s12859-019-3002-3", "venue_or_site": "BMC Bioinformatics", "abstract_or_summary": "Describes ROBOT, a command-line tool for automating ontology workflows such as reasoning, extraction, merging, reporting, quality checks, and release tasks.", "key_claims": [ "Ontology workflows can be automated through repeatable command-line operations.", "ROBOT supports reasoning, merging, extraction, validation reporting, and release workflows.", "Automated reports help make ontology quality issues visible and actionable." ], "ontology_relevance": "Key tooling source for ontology CI, release engineering, and machine-checkable quality practices.", "ai_relevance": "Automated ontology pipelines are needed when AI systems create, update, validate, and consume ontological knowledge at scale.", "palantir_relevance": "Useful comparator for enterprise ontology CI/CD and governance automation.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "automation", "biomedical-ontology", "ci-cd", "ontology-engineering", "ontology-workflows", "phase7", "robot", "validation" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase14-schemaorg-evolution-structured-data-web-2016", "title": "Schema.org: Evolution of Structured Data on the Web", "authors_or_org": "R. V. Guha; Dan Brickley; Steve Macbeth", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/2857274.2857276", "doi_or_identifier": "10.1145/2857274.2857276", "venue_or_site": "ACM Queue", "abstract_or_summary": "Schema.org documents the evolution of Web-scale structured data vocabularies used by publishers and consumers of structured data. The ACM Queue article traces the historical need for common schemas and explains Schema.org as a widely adopted vocabulary ecosystem for annotating Web content.", "key_claims": [ "Web-scale structured data requires common vocabularies that publishers and consumers can share.", "Schema.org demonstrates a pragmatic vocabulary ecosystem built for widespread publication and application use.", "Common schemas become more necessary as data volume and application diversity increase." ], "ontology_relevance": "Important lightweight-vocabulary comparator: not every useful ontology-adjacent system is a heavy OWL ontology; broad adoption can come from pragmatic shared schemas.", "ai_relevance": "Provides a canonical Web-scale example of shared vocabularies enabling machine interpretation, search, dataset discovery, and structured data consumption.", "palantir_relevance": "Useful contrast with Palantir's enterprise ontology: Schema.org is public, lightweight, Web-scale vocabulary infrastructure rather than a private operational ontology.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "common-vocabulary", "lightweight-ontology", "phase14", "schema-org", "semantic-web", "structured-data", "web-scale-semantics" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase6-oarga-2026-scientific-kg-ontology-open-llms", "title": "Scientific Knowledge Graph and Ontology Generation Using Open Large Language Models", "authors_or_org": "A. Oarga; M. Hart; A. M. Bran; M. Lederbauer; P. Schwaller", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://pubs.rsc.org/en/content/articlelanding/2026/dd/d5dd00275c", "doi_or_identifier": "10.1039/d5dd00275c", "venue_or_site": "Digital Discovery / Royal Society of Chemistry", "abstract_or_summary": "Uses open LLMs to generate scientific knowledge graphs and ontologies, then studies ontology-enhanced GraphRAG performance.", "key_claims": [ "Generated ontologies improve GraphRAG answers over KG-only baselines in the reported experiment.", "Open LLMs can help structure scientific literature into knowledge graphs and ontology hierarchies.", "Ontology inclusion improves comprehensiveness, diversity, and empowerment criteria in the reported evaluation." ], "ontology_relevance": "Excellent fit for ontology-guided scientific knowledge graphs and ontology-enhanced RAG pipelines.", "ai_relevance": "Links open LLM extraction, ontology generation, and GraphRAG evaluation.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "digital-discovery-2026", "graphrag", "knowledge-graph", "llm", "ontology-generation", "open-llm", "phase6", "scientific-kg" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-trist-bamforth-1951-sociotechnical-systems", "title": "Some Social and Psychological Consequences of the Longwall Method of Coal-Getting", "authors_or_org": "Eric L. Trist and K. W. Bamforth", "year": 1951, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1177/001872675100400101", "doi_or_identifier": "10.1177/001872675100400101", "venue_or_site": "Human Relations", "abstract_or_summary": "Foundational socio-technical systems study showing that technical redesign changes work organization, social relations, and human outcomes.", "key_claims": [ "Technical systems and social systems must be analyzed jointly.", "Efficiency-oriented redesign can damage work autonomy, coordination, and social organization.", "Optimization of a technical process can create organizational and psychological costs." ], "ontology_relevance": "Grounds the claim that ontology platforms must be evaluated as work systems, not only data structures.", "ai_relevance": "Agentic AI over enterprise workflows changes job roles, coordination, accountability, and autonomy.", "palantir_relevance": "Useful for assessing operational ontology and AIP as changes to work organization, not just analytics tooling.", "quality_signal": "widely_cited", "retrieval_tags": [ "critique", "human-factors", "organization", "phase3", "sociotechnical-systems", "work-design" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase3-star-ruhleder-1996-ecology-infrastructure", "title": "Steps Toward an Ecology of Infrastructure: Design and Access for Large Information Spaces", "authors_or_org": "Susan Leigh Star and Karen Ruhleder", "year": 1996, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1287/isre.7.1.111", "doi_or_identifier": "10.1287/isre.7.1.111", "venue_or_site": "Information Systems Research", "abstract_or_summary": "Classic study of large information spaces arguing that infrastructure emerges through relations among technology, work practice, access, and community.", "key_claims": [ "Infrastructure is not merely built; it becomes infrastructure in use.", "Access, membership, training, and local work practices shape whether shared systems succeed.", "Large information spaces fail when designers ignore local organizational realities." ], "ontology_relevance": "Supports a socio-technical view of ontology adoption: shared schemas must be sustained by work, access, and community.", "ai_relevance": "AI-ready ontologies require ongoing human practices of curation, permissioning, and interpretation.", "palantir_relevance": "Relevant to enterprise rollout where an ontology becomes useful only when embedded in operational routines.", "quality_signal": "widely_cited", "retrieval_tags": [ "access", "adoption", "information-infrastructure", "organizational-practice", "phase3", "sociotechnical" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "baader_2007_description_logic_handbook", "title": "The Description Logic Handbook: Theory, Implementation and Applications", "authors_or_org": "Franz Baader, Diego Calvanese, Deborah L. McGuinness, Daniele Nardi, and Peter F. Patel-Schneider, editors", "year": 2007, "source_type": "book", "bucket": "books", "url": "https://www.cambridge.org/core/books/description-logic-handbook/F050683766E57EE9BB07BC01BB7A7069", "doi_or_identifier": "isbn 9780521150118", "venue_or_site": "Cambridge University Press", "abstract_or_summary": "Canonical reference on description logics, the family of logics underlying OWL and many ontology reasoning systems.", "key_claims": [ "Description logics provide formal semantics for classes, roles, individuals, and subsumption.", "Decidability and complexity are central constraints on ontology language design.", "Reasoners can classify ontologies and detect inconsistency under formal semantics." ], "ontology_relevance": "Formal foundation for OWL-style ontologies and automated classification.", "ai_relevance": "Supports reliable symbolic reasoning and consistency checking in AI knowledge bases.", "palantir_relevance": "Useful where operational ontologies require class reasoning, constraint checking, or explainable type hierarchies.", "quality_signal": "widely_cited", "retrieval_tags": [ "description-logic", "formal-semantics", "owl", "reasoning" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase7-gene-ontology-knowledgebase-2023", "title": "The Gene Ontology knowledgebase in 2023", "authors_or_org": "The Gene Ontology Consortium", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/genetics/iyad031", "doi_or_identifier": "10.1093/genetics/iyad031", "venue_or_site": "GENETICS", "abstract_or_summary": "Updates the Gene Ontology knowledgebase and its annotation ecosystem, including GO-CAM model improvements, schema validation, annotation growth, historical archives, and cross-ontology consistency work.", "key_claims": [ "The Gene Ontology knowledgebase is a large computable resource for gene and gene-product function.", "GO-CAM extends term annotations into structured causal activity models.", "Schema validation and archival releases are necessary for reproducibility and traceability.", "Community ontology maintenance is an ongoing scientific infrastructure process." ], "ontology_relevance": "High-value Nature-facing evidence that ontology can operate as durable scientific infrastructure rather than only enterprise metadata.", "ai_relevance": "Shows a mature scientific ontology as computable training, annotation, retrieval, and validation infrastructure for biological AI systems.", "palantir_relevance": "Provides an independent non-commercial analogy for ontology as a shared operational layer for data, annotations, validation, and reuse.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biology", "biomedical-ontology", "gene-ontology", "go-cam", "ontology-infrastructure", "phase7", "scientific-infrastructure", "validation" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase15-ids-information-model-ontology-2020", "title": "The International Data Spaces Information Model - An Ontology for Sovereign Exchange of Digital Content", "authors_or_org": "Michael Schlueter Langdon; Sebastian Tramp; Jan Pennekamp; et al.; International Data Spaces contributors", "year": 2020, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/978-3-030-62466-8_12", "doi_or_identifier": "10.1007/978-3-030-62466-8_12", "venue_or_site": "International Semantic Web Conference", "abstract_or_summary": "The IDS Information Model paper presents an RDFS/OWL ontology defining the fundamental concepts of International Data Spaces for describing actors, resources, interactions, contracts, and data exchange in secure, sovereign, semantically interoperable data ecosystems.", "key_claims": [ "The IDS Information Model is an RDFS/OWL ontology for secure and semantically interoperable data spaces.", "It models actors, resources, interactions, contracts, and data exchange concepts needed for sovereign data sharing.", "Data spaces require semantic models in addition to protocols and connectors." ], "ontology_relevance": "Adds a cross-organization governance case where ontology mediates data sharing, contracts, and interoperability beyond a single enterprise boundary.", "ai_relevance": "Data-space architectures are relevant to agentic AI because they combine semantic interoperability, contracts, data sovereignty, and machine-readable ecosystem governance.", "palantir_relevance": "Useful comparator for Palantir public/enterprise platforms: data-space ontology highlights federation, sovereignty, and contract semantics rather than centralized platform control.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "data-sovereignty", "data-space", "dataspaces", "ids-information-model", "ontology", "phase15", "semantic-interoperability", "usage-policy" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "smith_etal_2007_obo_foundry", "title": "The OBO Foundry: Coordinated Evolution of Ontologies to Support Biomedical Data Integration", "authors_or_org": "Barry Smith, Michael Ashburner, Cornelius Rosse, et al.", "year": 2007, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/nbt1346", "doi_or_identifier": "10.1038/nbt1346", "venue_or_site": "Nature Biotechnology", "abstract_or_summary": "Sets out principles for a coordinated family of interoperable biomedical ontologies, including openness, common syntax, orthogonality, versioning, textual definitions, and shared relations.", "key_claims": [ "Ontology interoperability requires governance principles, not only shared technology.", "Orthogonality reduces overlap and inconsistency among ontology modules.", "Open, versioned, well-defined ontologies can support biomedical data integration." ], "ontology_relevance": "Major governance model for domain ontology ecosystems.", "ai_relevance": "Shows how ontologies can provide reusable, interoperable infrastructure for data-intensive AI.", "palantir_relevance": "Useful analogy for governing modular enterprise ontology domains with shared relations and versioning.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "governance", "interoperability", "obo-foundry", "versioning" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "guarino_oberle_staab_2009_what_is_ontology", "title": "What Is an Ontology?", "authors_or_org": "Nicola Guarino, Daniel Oberle, and Steffen Staab", "year": 2009, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/978-3-540-92673-3_0", "doi_or_identifier": "10.1007/978-3-540-92673-3_0", "venue_or_site": "Handbook on Ontologies, Springer", "abstract_or_summary": "Clarifies the meaning of ontology in philosophy and computer science and analyzes conceptualization, formal specification, and explicit commitment.", "key_claims": [ "The word ontology has distinct philosophical and computational meanings.", "Computational ontologies formalize aspects of a conceptualization for a purpose.", "Not every taxonomy, vocabulary, or graph has the same level of ontological commitment." ], "ontology_relevance": "High-value definitional source for avoiding category errors in ontology discourse.", "ai_relevance": "Helps distinguish AI knowledge graphs, schemas, vocabularies, embeddings, and formal ontologies.", "palantir_relevance": "Useful for scrutinizing whether an enterprise ontology is a formal commitment layer or a pragmatic object graph.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "conceptualization", "formal-ontology", "handbook", "ontology-definition" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "phase14-wikidata-free-collaborative-knowledgebase-2014", "title": "Wikidata: a free collaborative knowledgebase", "authors_or_org": "Denny Vrandečić; Markus Krötzsch", "year": 2014, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/2629489", "doi_or_identifier": "10.1145/2629489", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "The CACM article introduces Wikidata as a free collaborative knowledgebase for Wikipedia and broader reuse. It describes Wikidata's design choices, data model, multilingual structure, and role as an integrated source of factual data.", "key_claims": [ "Wikidata centralizes factual data for Wikipedia and wider reuse as a collaboratively edited knowledgebase.", "Its data model supports multilingual, structured statements that can serve many downstream applications.", "Open collaborative knowledgebases provide reusable entity and relation infrastructure for knowledge graph and AI systems." ], "ontology_relevance": "Adds a canonical open KG source for entity identity, schema reuse, community curation, and ontology-like structured statements.", "ai_relevance": "Wikidata is a major open knowledge graph substrate used for entity grounding, schema-constrained extraction, KG construction, and AI retrieval/evaluation.", "palantir_relevance": "Useful contrast with private enterprise ontologies: Wikidata shows public collaborative graph governance and open reuse, but with different quality-control and accountability tradeoffs.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "collaborative-knowledgebase", "entity-grounding", "knowledge-graph", "open-knowledge-graph", "phase14", "wikidata" ], "triage_tier": "candidate", "triage_score": 60 }, { "id": "oa-https-doi-org-10-1016-j-autcon-2020-103179", "title": "Towards a semantic Construction Digital Twin: Directions for future research", "authors_or_org": "Calin Boje, Annie Guerriero, S Kubicki, Yacine Rezgui", "year": 2020, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.autcon.2020.103179", "doi_or_identifier": "10.1016/j.autcon.2020.103179", "venue_or_site": "Automation in Construction", "abstract_or_summary": "As the Architecture, Engineering and Construction sector is embracing the digital age, the processes involved in the design, construction and operation of built assets are more and more influenced by technologies dealing with value-added monitoring of data from sensor networks, management of this data in secure and resilient storage systems underpinned by semantic models, as well as the simulation and optimisation of engineering systems. Aside from enhancing the efficiency of the value chain, such information-intensive models and associated technologies play a decisive role in minimising the lifecycle impacts of our buildings. While Building Information Modelling provides procedures, technologies and data schemas enabling a standardised semantic representation of building components and systems, the concept of a Digital Twin conveys a more holistic socio-technical and process-oriented characterisation of the complex artefacts involved by leveraging the synchronicity of the cyber-physical bi-directional data flows. Moreover, BIM lacks semantic completeness in areas such as control systems, including sensor networks, social systems, and urban artefacts beyond the scope of buildings, thus requiring a holistic, scalable semantic approach that factors in dynamic data at different levels. The paper reviews the multi-faceted applications of BIM during the construction stage and highlights limits and requirements, paving the way to the concept of a Construction Digital Twin. A definition of such a concept is then given, described in terms of underpinning research themes, while elaborating on areas for future research.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "bim", "construction", "dynamic-data", "foundational", "lifecycle", "ontology", "openalex", "semantic-digital-twin" ], "triage_tier": "candidate", "triage_score": 59 }, { "id": "oa-https-doi-org-10-1007-s00146-019-00932-9", "title": "Algorithms and values in justice and security", "authors_or_org": "Paul Hayes, Ibo van de Poel, Marc Steen", "year": 2020, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W2999513617", "doi_or_identifier": "10.1007/s00146-019-00932-9", "venue_or_site": "AI & Society", "abstract_or_summary": "Abstract This article presents a conceptual investigation into the value impacts and relations of algorithms in the domain of justice and security. As a conceptual investigation, it represents one step in a value sensitive design based methodology (not incorporated here are empirical and technical investigations). Here, we explicate and analyse the expression of values of accuracy, privacy, fairness and equality, property and ownership, and accountability and transparency in this context. We find that values are sensitive to disvalue if algorithms are designed, implemented or deployed inappropriately or without sufficient consideration for their value impacts, potentially resulting in problems including discrimination and constrained autonomy. Furthermore, we outline a framework of conceptual relations of values indicated by our analysis, and potential value tensions in their implementation and deployment with a view towards supporting future research, and supporting the value sensitive design of algorithms in justice and security.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase3-llm-ontmem-017", "title": "AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents", "authors_or_org": "Petr Anokhin; Nikita Semenov; Artyom Sorokin; Dmitry Evseev; Andrey Kravchenko; Mikhail Burtsev; Evgeny Burnaev", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://www.ijcai.org/proceedings/2025/0002.pdf", "doi_or_identifier": "ijcai:2025:0002", "venue_or_site": "IJCAI 2025", "abstract_or_summary": "Presents AriGraph, a memory architecture in which an LLM agent builds and updates a knowledge graph world model integrating semantic and episodic memories for planning in interactive environments.", "key_claims": [ "Unstructured memory summaries are weak substrates for planning in complex environments.", "Knowledge graph world models can combine semantic and episodic memory for agent decision-making.", "Dynamic graph memory improves performance in interactive text game environments and multi-hop QA." ], "ontology_relevance": "Strong peer-reviewed evidence for graph-shaped agent memory as an ontology-like world model.", "ai_relevance": "Connects LLM agents, planning, episodic memory, semantic memory, and KG updates.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "agent-memory", "arigraph", "episodic-memory", "ijcai-2025", "knowledge-graph-world-model", "llm-agents", "planning" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "industry-commercial-jupiterone-neptune", "title": "JupiterOne security graph case materials on AWS Neptune", "authors_or_org": "Amazon Web Services / JupiterOne", "year": 2022, "source_type": "case_study", "bucket": "commercial", "url": "https://aws.amazon.com/solutions/case-studies/jupiterone", "doi_or_identifier": null, "venue_or_site": "AWS Case Studies", "abstract_or_summary": "AWS and JupiterOne materials describe a security knowledge graph that connects assets, identities, vulnerabilities, and policies for cyber asset management and security analysis.", "key_claims": [ "Security posture can be represented as connected assets, identities, policies, vulnerabilities, and findings.", "Graph queries support exposure path and blast-radius analysis.", "Many external integrations can feed a near-current operational graph." ], "ontology_relevance": "Good example of an ontology-like asset and relationship model used in production security workflows.", "ai_relevance": "Security graphs can provide structured context for AI incident response, risk explanation, and compliance assistance.", "palantir_relevance": "Comparable to operational ontology for cyber and asset intelligence outside Palantir.", "quality_signal": "primary_source", "retrieval_tags": [ "aws", "case-study", "cybersecurity", "jupiterone", "neptune", "security-graph" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase3-llm-ontmem-007", "title": "Large Language Models as Assistants for Ontology Engineering", "authors_or_org": "Javad Saeedizade", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://ceur-ws.org/Vol-4085/paper18.pdf", "doi_or_identifier": "ceur-ws vol-4085 paper18", "venue_or_site": "ISWC 2025 Companion / CEUR-WS", "abstract_or_summary": "PhD research proposal for an LLM-based ontology engineering assistant combining conceptual modeling suggestions, pattern-based class and property definitions, reasoner checks, and OOPS-style pitfall detection.", "key_claims": [ "Ontology engineering assistants should combine generation with validation, not only provide text or OWL drafts.", "Reasoners and ontology pitfall scanners are natural guardrails for LLM-assisted modeling.", "Pattern-based suggestions can reduce dependence on scarce ontology expertise while preserving modeling discipline." ], "ontology_relevance": "Useful forward-looking source for integrated LLM ontology tooling that includes validation loops.", "ai_relevance": "Positions LLM assistance inside a toolchain with symbolic checks and evaluation components.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "design-patterns", "llm-assistant", "ontology-engineering", "oops", "reasoner" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "n3c-ncats-faqs-2026", "title": "National COVID Cohort Collaborative (N3C) frequently asked questions", "authors_or_org": "National Center for Advancing Translational Sciences, NIH", "year": 2026, "source_type": "webpage", "bucket": "academic", "url": "https://ncats.nih.gov/research/research-activities/n3c/faqs", "doi_or_identifier": null, "venue_or_site": "NCATS, National Institutes of Health", "abstract_or_summary": "NIH/NCATS FAQ for N3C, describing the COVID research data enclave, participating institutions, data protections, access controls, and governance structure.", "key_claims": [ "N3C is described as a centralized, secure data enclave for COVID-related clinical research data.", "Access, data contribution, and governance are managed through institutional and NIH processes.", "The FAQ provides official public context for a major health-data enclave associated in the literature with Palantir Foundry infrastructure.", "N3C data use is limited to approved clinical and translational research questions.", "Data-use access is governed by Data Use Requests and agreements.", "Oversight committee governance and disclosure restrictions are part of the access model." ], "ontology_relevance": "Public example of large-scale governed clinical data integration, relevant to operational ontology/data-platform design.", "ai_relevance": "N3C's analytics-ready data enclave is a precedent for governed AI/ML research environments.", "palantir_relevance": "Context source for a public health-data platform where Palantir Foundry has been publicly discussed in related N3C materials and literature.", "quality_signal": "primary_source", "retrieval_tags": [ "covid", "data-enclave", "dua", "governance", "health-data", "health-data-governance", "n3c", "ncats", "nih" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "uk-hansard-nhs-fdp-2026", "title": "NHS Federated Data Platform debate", "authors_or_org": "UK Parliament Hansard", "year": 2026, "source_type": "other", "bucket": "technical", "url": "https://hansard.parliament.uk/commons/2026-04-16/debates/2FDCA71C-D0C1-4738-BEE8-A4BDA311DB99/NHSFederatedDataPlatform", "doi_or_identifier": "hansard debate 16 april 2026", "venue_or_site": "Hansard, UK Parliament", "abstract_or_summary": "Parliamentary debate record on the NHS Federated Data Platform, including concerns about value, claims, auditability, and Palantir's role.", "key_claims": [ "Parliamentary participants questioned Palantir/NHS benefit claims and value for money.", "The debate noted that the National Audit Office had not yet assessed some delivery-value claims.", "The debate shows political scrutiny of the FDP.", "MPs raised concerns about transparency, privacy, public trust, and Palantir's involvement in NHS data infrastructure.", "Hansard provides accountable public-sector scrutiny rather than vendor or campaigner framing alone.", "The debate is useful for mapping governance concerns around operational data platforms.", "MPs argued that heavy redaction of the Palantir FDP contract makes scrutiny difficult.", "The debate frames FDP governance as a political and public-accountability issue, not only a technical matter.", "Parliamentary scrutiny centers on transparency, consent, and public-sector accountability." ], "ontology_relevance": "Governance evidence for large-scale public ontology/data-platform deployments.", "ai_relevance": "Relevant to accountability of AI-ready public infrastructure.", "palantir_relevance": "Primary public record of political scrutiny concerning Palantir's NHS role.", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "governance", "hansard", "nhs", "nhs-fdp", "palantir", "public-accountability", "public-scrutiny", "scrutiny", "transparency", "uk-parliament" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "nhs-fdp-faqs-2026", "title": "NHS Federated Data Platform: Frequently asked questions", "authors_or_org": "NHS England", "year": 2026, "source_type": "webpage", "bucket": "technical", "url": "https://www.england.nhs.uk/digitaltechnology/nhs-federated-data-platform/fdp-faqs", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "Official NHS FAQ explaining the Federated Data Platform, including that it is built on Palantir Foundry and describing contractual/privacy safeguards and IP arrangements.", "key_claims": [ "The NHS Federated Data Platform is built on Palantir Foundry software.", "NHS England says Palantir retains IP rights in Foundry but NHS interests are protected contractually.", "NHS England says benefits statistics are updated quarterly.", "NHS England presents the FDP as a platform for connecting NHS data to improve planning and care delivery.", "The FAQ identifies Palantir as part of the supplier group and addresses public concerns about data use.", "The source is official NHS communication and should be read alongside contract, privacy, and parliamentary scrutiny materials." ], "ontology_relevance": "Official public-sector example of Foundry as shared operational data platform.", "ai_relevance": "Useful for governance questions around AI-ready public data infrastructure.", "palantir_relevance": "Primary non-Palantir official source for a major Palantir deployment.", "quality_signal": "primary_source", "retrieval_tags": [ "contract", "federated-data-platform", "governance", "health-data", "nhs", "palantir", "palantir-foundry", "privacy", "public-sector" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase3-nhs-fdp-official-overview-2026", "title": "NHS Federated Data Platform: official overview and FAQs", "authors_or_org": "NHS England", "year": 2026, "source_type": "webpage", "bucket": "synthesis", "url": "https://www.england.nhs.uk/digitaltechnology/nhs-federated-data-platform", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "Official NHS England overview for the Federated Data Platform, explaining its intended role in connecting data for planning, care coordination, operational improvement, and local/national NHS use cases.", "key_claims": [ "The FDP is framed as infrastructure to connect information held in separate NHS systems.", "NHS England presents the platform as supporting operational planning and care improvement.", "The official framing emphasizes NHS purposes and governance, not independent evaluation." ], "ontology_relevance": "A public-sector data-integration case where ontology-like operational models can become infrastructure.", "ai_relevance": "Provides context for AI-enabled health operations built on connected data platforms.", "palantir_relevance": "Primary deployment context for Palantir software in the UK public sector.", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "governance", "health-data", "nhs", "official-overview", "public-sector" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase4-std-omg-odm-1-1-2014", "title": "Ontology Definition Metamodel, Version 1.1", "authors_or_org": "Object Management Group", "year": 2014, "source_type": "standard", "bucket": "technical", "url": "https://www.omg.org/spec/ODM/1.1", "doi_or_identifier": null, "venue_or_site": "OMG Formal Specification", "abstract_or_summary": "OMG specification for ontology definition and exchange across modeling languages, including metamodels and mappings for RDF, OWL, Topic Maps, Common Logic, and UML profiles.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "metamodel", "model-interchange", "odm", "omg", "ontology", "semantic-interoperability", "uml" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase2-graphrag-009", "title": "OpenSPG KAG: Knowledge Augmented Generation framework", "authors_or_org": "OpenSPG / Ant Group community", "year": 2024, "source_type": "docs", "bucket": "technical", "url": "https://github.com/OpenSPG/KAG", "doi_or_identifier": "github:openspg/kag", "venue_or_site": "GitHub", "abstract_or_summary": "Official open-source KAG framework and examples for building domain knowledge bases, knowledge alignment, hybrid retrieval, and reasoning-augmented generation.", "key_claims": [ "KAG implements retrieval over knowledge, chunks, and schema-like graph structures.", "Open-source artifacts make professional-domain KAG reproducible beyond a paper claim.", "Schema, alignment, and reasoning are treated as engineering components, not only prompts." ], "ontology_relevance": "Concrete implementation reference for ontology-like schema-driven retrieval and graph-grounded generation.", "ai_relevance": "Useful for evaluating how KAG differs from GraphRAG and vector-only RAG in practice.", "palantir_relevance": "Comparable to ontology-centric enterprise AI stacks because it couples domain graph construction with answer generation.", "quality_signal": "official_open_source", "retrieval_tags": [ "kag", "knowledge-base", "open-source", "openspg", "reasoning", "schema-retrieval" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase2-graphrag-014", "title": "Retrieval-Augmented Generation with Graphs (GraphRAG)", "authors_or_org": "Haoyu Han; Yu Wang; Harry Shomer; Kai Guo; Jiayuan Ding; Yongjia Lei; Mahantesh Halappanavar; Ryan A. Rossi; Subhabrata Mukherjee; Xianfeng Tang; Qi He; Zhigang Hua; Bo Long; Tong Zhao; Neil Shah; Amin Javari; Yinglong Xia; Jiliang Tang", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2501.00309", "doi_or_identifier": "arxiv:2501.00309", "venue_or_site": "arXiv", "abstract_or_summary": "Survey-style treatment of GraphRAG systems, organizing graph construction, graph retrieval, augmentation, generation, and applications for LLMs.", "key_claims": [ "GraphRAG should be evaluated as a pipeline spanning graph construction, retrieval, and generation.", "Graph structure can improve context organization, multi-hop retrieval, and explainability.", "Noisy graph construction can erase the gains promised by graph-aware retrieval." ], "ontology_relevance": "Useful source for distinguishing formal ontology-backed GraphRAG from ad hoc generated graph indexes.", "ai_relevance": "Current synthesis of graph retrieval-augmented generation design patterns.", "palantir_relevance": "Supports comparing curated operational ontologies with generated GraphRAG pipelines.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "graph-construction", "graph-retrieval", "graphrag", "llm", "survey" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "phase3-llm-ontmem-011", "title": "Schema Matching with Large Language Models: An Experimental Study", "authors_or_org": "Marcel Parciak; Brecht Vandevoort; Frank Neven; Liesbet M. Peeters; Stijn Vansummeren", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2407.11852", "doi_or_identifier": "arxiv:2407.11852", "venue_or_site": "VLDBW 2024 / TaDA 2024", "abstract_or_summary": "Experimental study of LLMs for schema matching, closely related to ontology matching, evaluating how large language models behave on data integration matching tasks.", "key_claims": [ "LLMs are useful for some schema matching settings but their gains depend strongly on task structure and available metadata.", "Schema matching is a practical proxy for ontology and semantic layer alignment in enterprise data integration.", "Evaluation must separate lexical matches from genuinely semantic correspondences." ], "ontology_relevance": "Connects ontology alignment to broader schema matching and semantic data integration.", "ai_relevance": "Tests LLM semantic matching capabilities in a structured data integration setting.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "data-integration", "llm", "schema-matching", "semantic-alignment", "vldbw-2024" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "industry-commercial-stardog-cases", "title": "Stardog enterprise knowledge graph customer case studies", "authors_or_org": "Stardog", "year": 2026, "source_type": "case_study", "bucket": "commercial", "url": "https://www.stardog.com/customers", "doi_or_identifier": null, "venue_or_site": "Stardog Customer Stories", "abstract_or_summary": "Stardog customer stories describe knowledge graph applications in scientific publishing, aerospace, energy, manufacturing, and other enterprise domains.", "key_claims": [ "Enterprise KGs can integrate specialized domain data across legacy systems and expert datasets.", "Ontology can normalize terminology and relationships for discovery and reuse.", "Vendor case studies often provide architecture motifs but limited independent measurement." ], "ontology_relevance": "Useful examples of semantic integration and ontology-backed discovery across complex domains.", "ai_relevance": "Curated graph data can be used as context for AI search and question answering.", "palantir_relevance": "Shows similar semantic integration ambitions outside Palantir's platform ecosystem.", "quality_signal": "primary_source", "retrieval_tags": [ "case-studies", "enterprise-kg", "scientific-data", "semantic-integration", "stardog" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "s2-10-3390-info16050365", "title": "Toward Robust Security Orchestration and Automated Response in Security Operations Centers with a Hyper-Automation Approach Using Agentic Artificial Intelligence", "authors_or_org": "Ismail, Rahmat Kurnia, Zilmas Arjuna Brata, Ghitha Afina Nelistiani, Shinwook Heo, Hyeong-Sik Kim, Howon Kim", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/8e9df72ad7b93769f3384fb746bbd53a8dc66fb3", "doi_or_identifier": "10.3390/info16050365", "venue_or_site": "Inf.", "abstract_or_summary": "The evolving landscape of cybersecurity threats demands the modernization of Security Operations Centers (SOCs) to enhance threat detection, response, and mitigation. Security Orchestration, Automation, and Response (SOAR) platforms play a crucial role in addressing operational inefficiencies; however, traditional no-code SOAR solutions face significant limitations, including restricted flexibility, scalability challenges, inadequate support for advanced logic, and difficulties in managing large playbooks. These constraints hinder effective automation, reduce adaptability, and underutilize analysts’ technical expertise, underscoring the need for more sophisticated solutions. To address these challenges, we propose a hyper-automation SOAR platform powered by agentic-LLM, leveraging Large Language Models (LLMs) to optimize automation workflows. This approach shifts from rigid no-code playbooks to AI-generated code, providing a more flexible and scalable alternative while reducing operational complexity. Additionally, we introduce the IVAM framework, comprising three critical stages: (1) Investigation, structuring incident response into actionable steps based on tailored recommendations, (2) Validation, ensuring the accuracy and effectiveness of executed actions, (3) Active Monitoring, providing continuous oversight. By integrating AI-driven automation with the IVAM framework, our solution enhances investigation quality, improves response accuracy, and increases SOC efficiency in addressing modern cybersecurity threats.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 58 }, { "id": "oa-https-openalex-org-w1563421732", "title": "Ontological Engineering: with examples from the areas of Knowledge Management, e-Commerce and the Semantic Web", "authors_or_org": "Asuncíon Gómez-Pérez, Mariano Fernández‐López, Óscar Corcho", "year": 2004, "source_type": "paper", "bucket": "academic", "url": "https://openalex.org/W1563421732", "doi_or_identifier": null, "venue_or_site": "", "abstract_or_summary": "Ontological Engineering refers to the set of activities that concern the ontology development process, the ontology life cycle, the methods and methodologies for building ontologies, and the tool suites and languages that support them. \\n \\nDuring the last decade, increasing attention has been focused on ontologies and Ontological Engineering. Ontologies are now widely used in Knowledge Engineering, Artificial Intelligence and Computer Science; in applications related to knowledge management, natural language processing, e-commerce, intelligent integration information, information retrieval, integration of databases, bioinformatics, and education; and in new emerging fields like the Semantic Web. \\n \\nPrimary goals of this book are to acquaint students, researchers and developers of information systems with the basic concepts and major issues of Ontological Engineering, as well as to make ontologies more understandable to those computer science engineers that integrate ontologies into their information systems. We have paid special attention to the influence that ontologies have on the Semantic Web. Pointers to the Semantic Web appear in all the chapters, but specially in the chapter on ontology languages and tools. \\n \\nMany different methods, tools and languages, as well as the most out standing ontologies, are presented to illustrate a diversity of approaches, but no single technique receives special attention. Each individual may choose to emphasize particular techniques depending on his/her own circumstances and interests. So, the book is designed to operate at two levels. First, as a simple introduction to the major areas of Ontological Engineering, and second, as a reference book. The emerging areas and the most up-to-date standards have also been considered. \\n \\nThe layout of the text is divided into five chapters: theoretical foundations, the most outstanding ontologies, methodologies, languages, and tools for developing ontologies. In every chapter (except the chapter that describes the most out standing ontologies) we have used examples taken from the traveling domain. This provides a focal point for the book and allows readers to practise and compare different modeling techniques, different and similar methods and methodologies for building ontologies, and learn about ontology languages and different types of tools. We also include comparative studies of methodologies, tools and languages to advise ontologists on their use. \\n \\nThe first chapter contains the theoretical foundations of the ontology field. Here we explain what an ontology is, the main types of ontologies, the main modeling components of ontologies based on frames or description logic, the design criteria for building ontologies as well as the relationships with other modelling techniques that are widely used on software engineering and databases. \\n \\nChapter 2 is devoted to the most outstanding ontologies. We present different types of ontologies: knowledge representation ontologies of traditional (i.e., On to lingua and OKBC) and ontology mark-up languages (i.e., RDF(S), OIL,DAML+OIL, and OWL), top level ontologies, linguistic ontologies, and domain ontologies in the areas of e-commerce, medicine, engineering, enterprise, chemistry and knowledge management. \\n \\nIn Chapter 3 we explore different methods and methodologies for ontology construction. We present in detail the ontology development process and the methods and methodologies that support the ontology construction from scratch. We also discuss particular methods that allow specific activities. Special attention is given to the ontology learning methods that reduce the effort during the knowledge acquisition process; the merging of ontologies that generates a unique ontology from several ontologies; the ontology alignment that establishes different types of mapping between ontologies (hence preserving the original ones); and the ontology evaluation for evaluating the ontology content. For each methodology and method, we give an example taken from the traveling domain. \\n \\nChapter 4 deals with the process of selecting the ontology language (or set of languages) in which the ontology will be implemented. We describe how to implement ontologies in classical languages (On to lingua, KIF, OCML and FLogic), the OKBC protocol, and web-based ontology languages (SHOE, XOL, RDF(S),OIL, DAML+OIL and OWL) that have laid the foundations of the Semantic Web. Some of them, like RDF(S) and OWL, are still in a development phase. As we have implemented an ontology of the traveling domain in all these languages, we compare their expressiveness and the reasoning mechanisms of each language. \\n \\nFinally, Chapter 5 is concerned with several types of the tools and platforms used to build ontologies and tools that allow the use of ontologies for the Semantic Web. As in the previous chapters, we provide examples of ontologies with the tools of the traveling domain.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "foundational", "openalex" ], "triage_tier": "candidate", "triage_score": 57 }, { "id": "phase4-nature-kejriwal-2021-kg-fundamentals", "title": "Knowledge Graphs: Fundamentals, Techniques, and Applications", "authors_or_org": "Mayank Kejriwal, Craig A. Knoblock, Pedro Szekely", "year": 2021, "source_type": "book", "bucket": "books", "url": "https://mitpress.mit.edu/9780262045094/knowledge-graphs", "doi_or_identifier": null, "venue_or_site": "MIT Press", "abstract_or_summary": "Covers knowledge graph construction, entity linkage, information extraction, schema design, and applications across heterogeneous data environments.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_book", "retrieval_tags": [ "applications", "construction", "data-integration", "entity-linking", "knowledge-graph" ], "triage_tier": "candidate", "triage_score": 56 }, { "id": "p2-llm-ont-008", "title": "A RAG Approach for Generating Competency Questions in Ontology Engineering", "authors_or_org": "Xueli Pan; Jacco van Ossenbruggen; Victor de Boer; Zhisheng Huang", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2409.08820", "doi_or_identifier": "arxiv:2409.08820", "venue_or_site": "MTSR 2024 / arXiv", "abstract_or_summary": "Uses retrieval-augmented generation over scientific papers to generate competency questions for ontology engineering and compares against expert-authored questions.", "key_claims": [ "Domain retrieval improves competency question generation over zero-shot prompting.", "Precision and consistency are useful but incomplete evaluation signals for generated CQs.", "RAG can help elicit ontology requirements before formal modeling begins." ], "ontology_relevance": "Targets the requirements stage of ontology engineering, not only extraction or population.", "ai_relevance": "Shows RAG as a knowledge elicitation tool for LLM-supported modeling.", "palantir_relevance": "Useful for turning enterprise documents into candidate business questions that an ontology should answer.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "competency-questions", "llm", "mtsr-2024", "ontology-requirements", "rag" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "ont-ai-014", "title": "A Survey on Augmenting Knowledge Graphs (KGs) with Large Language Models (LLMs): Models, Evaluation Metrics, Benchmarks, and Challenges", "authors_or_org": "Nourhan Ibrahim; Samar Aboulela; Ahmed Ibrahim; Rasha Kashef", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://link.springer.com/article/10.1007/s44163-024-00175-8", "doi_or_identifier": "10.1007/s44163-024-00175-8", "venue_or_site": "Discover Artificial Intelligence", "abstract_or_summary": "Survey of approaches that augment knowledge graph construction, completion, reasoning, and applications with LLM capabilities, including evaluation metrics and challenges.", "key_claims": [ "LLMs can support KG construction, completion, reasoning, and natural-language interfaces.", "KG-LLM systems require evaluation across model behavior, graph quality, benchmarks, and task outcomes.", "Hybrid LLM-KG pipelines face challenges in factuality, scalability, and verification." ], "ontology_relevance": "Relevant to ontology-backed KG lifecycle and query interfaces.", "ai_relevance": "Summarizes how LLMs are applied to structured graph knowledge.", "palantir_relevance": "Relevant to natural-language interfaces for enterprise object graphs.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "kg-construction", "knowledge-graph", "llm", "reasoning", "survey" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase2-ares-2024", "title": "ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems", "authors_or_org": "Jon Saad-Falcon; Omar Khattab; Christopher Potts; Matei Zaharia", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2311.09476", "doi_or_identifier": "arxiv:2311.09476", "venue_or_site": "NAACL 2024 / arXiv", "abstract_or_summary": "Automated RAG evaluation system that creates synthetic data, fine-tunes lightweight judges, and evaluates context relevance, answer faithfulness, and answer relevance with prediction-powered inference.", "key_claims": [ "Automated RAG evaluation can be made more reliable by training task-specific judges and using small human-labeled sets for calibration.", "RAG evaluation should distinguish retrieval quality from answer faithfulness and answer relevance.", "Confidence intervals and prediction-powered inference help quantify evaluation uncertainty.", "RAG systems need scalable evaluation without exhaustive human labeling.", "Synthetic queries and trained judges can support domain-specific RAG evaluation.", "Evaluation should distinguish retrieval failure from generation failure." ], "ontology_relevance": "Useful for evaluating ontology-based retrieval and generated claims with calibrated metrics.", "ai_relevance": "Important source for scalable RAG evaluation and trustworthy deployment.", "palantir_relevance": "Provides a general evaluation framework to compare with Palantir's AIP Evals claims.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ares", "evaluation", "faithfulness", "llm-judge", "ppi", "rag", "rag-evaluation", "retrieval-failure", "synthetic-queries", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-crawford-2021-atlas-ai", "title": "Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence", "authors_or_org": "Kate Crawford", "year": 2021, "source_type": "book", "bucket": "books", "url": "https://yalebooks.yale.edu/book/9780300264630/atlas-of-ai", "doi_or_identifier": "isbn 9780300264630", "venue_or_site": "Yale University Press", "abstract_or_summary": "Critical book tracing AI through labor, data extraction, classification, environmental costs, state power, and political economy.", "key_claims": [ "AI systems depend on material infrastructures, labor, data extraction, and institutional power.", "Classification is a central mechanism through which AI systems order people and the world.", "AI cannot be evaluated only as software; it must be analyzed as political economy and infrastructure." ], "ontology_relevance": "Strong critique source for ontology as classification and infrastructure embedded in power relations.", "ai_relevance": "Useful counterweight to purely technical AI governance accounts, especially for large-scale platform deployment.", "palantir_relevance": "Relevant to analyzing military, government, and enterprise AI infrastructures as institutional power systems.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-critique", "classification", "infrastructure", "labor", "phase3", "political-economy" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase11-qian-2026-brick-dicl-schema-classification", "title": "Brick-DICL: Dynamic In-Context Learning for Automated Brick Schema Classification", "authors_or_org": "Yiyue Qian; Shinan Zhang; Huan Song; Negin Sokhandan; Hannah Marlowe; Diego Socolinsky", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.17637", "doi_or_identifier": "10.48550/arxiv.2606.17637", "venue_or_site": "arXiv", "abstract_or_summary": "Proposes Brick-DICL, a dynamic in-context learning framework for automated Brick schema classification in building management systems. It combines metadata-RAG, class-RAG, and multi-LLM filtering to classify heterogeneous BMS points against a standardized building ontology while flagging uncertain cases for human review.", "key_claims": [ "Brick schema provides a standardized ontology for building systems, but mapping heterogeneous BMS points to classes is difficult.", "Metadata-RAG and class-RAG are used to improve LLM domain knowledge and reduce the large class search space.", "Multi-LLM filtering flags low-confidence classifications for human review.", "The framework targets faster interoperable digital building onboarding while keeping manual verification in the loop." ], "ontology_relevance": "Shows ontology-grounded classification as an applied enterprise/industrial pattern: domain ontology narrows and validates LLM outputs.", "ai_relevance": "Current applied evidence that domain ontologies can structure LLM-assisted classification and onboarding in industrial/building data systems.", "palantir_relevance": "Non-Palantir operational comparator for mapping messy operational assets into a semantic model.", "quality_signal": "preprint_benchmark", "retrieval_tags": [ "brick-schema", "building-management-system", "class-rag", "domain-ontology", "human-in-the-loop", "llm", "metadata-rag", "ontology-classification", "phase11", "semantic-interoperability" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-raji-etal-2020-accountability-gap", "title": "Closing the AI Accountability Gap: Defining an End-to-End Framework for Internal Algorithmic Auditing", "authors_or_org": "Inioluwa Deborah Raji, Andrew Smart, Rebecca N. White, Margaret Mitchell, Timnit Gebru, Ben Hutchinson, Jamila Smith-Loud, Daniel Theron, and Parker Barnes", "year": 2020, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3351095.3372873", "doi_or_identifier": "10.1145/3351095.3372873", "venue_or_site": "ACM Conference on Fairness, Accountability, and Transparency", "abstract_or_summary": "Paper proposing internal algorithmic auditing as an end-to-end governance process across AI development and deployment.", "key_claims": [ "AI accountability requires process controls, documentation, review, and organizational responsibility.", "Auditing should cover the full lifecycle, not only model performance at one point in time.", "Accountability gaps arise when responsibility is fragmented across teams and stages." ], "ontology_relevance": "Provides governance process vocabulary for ontology-backed AI lifecycle controls and change review.", "ai_relevance": "Directly relevant to production AI governance, auditability, documentation, and organizational accountability.", "palantir_relevance": "Useful for evaluating platform claims about observability, evals, permissions, and audit trails.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "accountability", "ai-audit", "documentation", "lifecycle-governance", "organizational-controls", "phase3" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-gebru-etal-2021-datasheets", "title": "Datasheets for Datasets", "authors_or_org": "Timnit Gebru, Jamie Morgenstern, Briana Vecchione, Jennifer Wortman Vaughan, Hanna Wallach, Hal Daume III, and Kate Crawford", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3458723", "doi_or_identifier": "10.1145/3458723", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "Influential proposal for standardized dataset documentation covering motivation, composition, collection, preprocessing, uses, distribution, and maintenance.", "key_claims": [ "Dataset documentation can reduce misuse and improve transparency.", "Datasets need documentation of collection context, composition, preprocessing, recommended uses, and maintenance.", "Documentation helps expose limitations, biases, and responsibilities." ], "ontology_relevance": "Provides a template for documenting ontology-backed datasets, mappings, provenance, and intended use.", "ai_relevance": "AI systems grounded in knowledge graphs require documented data sources, transformations, and limitations.", "palantir_relevance": "Relevant to documenting enterprise object models and data products that feed AI workflows.", "quality_signal": "widely_cited", "retrieval_tags": [ "data-governance", "dataset-documentation", "datasheets", "maintenance", "phase3", "transparency" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase7-edam-bioinformatics-ontology-2013", "title": "EDAM: an ontology of bioinformatics operations, types of data and identifiers, topics and formats", "authors_or_org": "Ison; Kalaš; Jonassen; et al.", "year": 2013, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/bioinformatics/btt113", "doi_or_identifier": "10.1093/bioinformatics/btt113", "venue_or_site": "Bioinformatics", "abstract_or_summary": "EDAM defines an ontology for bioinformatics operations, data types, identifiers, topics, and formats, supporting tool annotation, workflow composition, and data interoperability.", "key_claims": [ "Bioinformatics tools and workflows benefit from shared ontology terms for operations, data, identifiers, topics, and formats.", "Tool annotation becomes more interoperable when inputs and outputs are semantically typed.", "Ontology supports workflow discovery and composition." ], "ontology_relevance": "Important bridge between scientific ontology and tool/action semantics.", "ai_relevance": "Shows how ontology can describe computational tools, inputs, outputs, and formats, a direct parallel to agent tool schemas and model-context protocols.", "palantir_relevance": "Direct comparator for Palantir Ontology MCP and action/function exposure: tools need typed semantic descriptions.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "bioinformatics", "edam", "ontology-infrastructure", "phase7", "scientific-workflows", "tool-semantics", "workflow" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase16-fairsharing-community-standards-repositories-policies-2019", "title": "FAIRsharing as a Community Approach to Standards, Repositories and Policies", "authors_or_org": "Allyson L. Lister; Susanna-Assunta Sansone; et al.", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/s41587-019-0080-8", "doi_or_identifier": "10.1038/s41587-019-0080-8", "venue_or_site": "Nature Biotechnology", "abstract_or_summary": "FAIRsharing describes a community platform for standards, repositories, and data policies. It catalogs and connects standards, databases, and policies so researchers and systems can discover appropriate metadata, reporting, and identifier resources.", "key_claims": [ "Scientific data reuse depends on discoverable standards, repositories, and policies.", "Registries help communities converge on shared reporting and metadata resources.", "Agents that reason over scientific data need machine-discoverable standards and repository context." ], "ontology_relevance": "Strengthens the open standards and curated knowledge infrastructure route with a standards/repository registry source.", "ai_relevance": "Ontology-backed AI and scientific RAG systems need registries of standards and repositories so agents can select appropriate schemas, vocabularies, and policy constraints rather than inventing local conventions.", "palantir_relevance": "Neutral comparator for enterprise catalogs and governance registries, but not Palantir-specific evidence.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "curated-knowledge-infrastructure", "fair-data", "fairsharing", "metadata-standards", "ontology-governance", "phase16", "repositories", "standards-registry" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase2-doumanas-2025-finetuning-ontology-engineering", "title": "Fine-Tuning Large Language Models for Ontology Engineering: A Comparative Analysis of GPT-4 and Mistral", "authors_or_org": "Dimitrios Doumanas; Andreas Soularidis; Dimitris Spiliotopoulos; Costas Vassilakis; Konstantinos Kotis", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://www.mdpi.com/2076-3417/15/4/2146", "doi_or_identifier": "10.3390/app15042146", "venue_or_site": "Applied Sciences", "abstract_or_summary": "Compares fine-tuning and model choices for LLM-assisted ontology engineering tasks using GPT-4 and Mistral-family models.", "key_claims": [ "Model adaptation can matter for ontology engineering quality, especially in domain-specific tasks.", "Prompt-only and fine-tuned approaches should be compared against explicit ontology quality criteria.", "Open and proprietary models may differ in cost, controllability, and performance for ontology workflows." ], "ontology_relevance": "Adds model-adaptation evidence to the LLM ontology engineering literature.", "ai_relevance": "Useful for discussing whether ontology work should rely on general LLMs, fine-tuned LLMs, or hybrid workflows.", "palantir_relevance": "Relevant to Palantir AIP's multi-model and enterprise workflow positioning.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "fine-tuning", "llm", "model-comparison", "ontology-engineering", "validation" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase11-shuai-2026-usd-scenes-ontology-grounding-llms", "title": "From USD Scenes to Knowledge Graphs: Zero-Shot Ontology Grounding with LLMs", "authors_or_org": "Jiangtao Shuai; Zongxiong Chen; Manfred Hauswirth; Sonja Schimmler", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.09134", "doi_or_identifier": "10.48550/arxiv.2606.09134", "venue_or_site": "arXiv", "abstract_or_summary": "Studies zero-shot ontology grounding from Universal Scene Description scenes to formal ontology classes using LLMs. The reported experiments show high exact-match accuracy when semantic scene cues are available and sharp degradation when cues are anonymized, highlighting both promise and dependence on meaningful context.", "key_claims": [ "LLMs can ground scene objects to formal ontology classes in zero-shot settings when descriptive semantic cues are available.", "Scene-graph context such as sibling names and parent paths is a major driver of grounding performance.", "Opaque names and anonymized semantic cues sharply reduce performance, showing limits of LLM grounding without meaningful context.", "Ontology grounding is a bottleneck for constructing KGs from 3D simulation scenes for robot task reasoning." ], "ontology_relevance": "Extends ontology+AI coverage beyond text and enterprise documents into embodied scene grounding and KG construction.", "ai_relevance": "Current evidence for LLM-assisted ontology grounding in embodied/robotic scene understanding, including limits when semantic cues are removed.", "palantir_relevance": "Indirect comparator for operational digital twins and simulation environments; not Palantir-specific evidence.", "quality_signal": "preprint_benchmark", "retrieval_tags": [ "digital-twin", "embodied-ai", "knowledge-graph", "llm", "ontology-grounding", "phase11", "robot-task-reasoning", "scene-graph", "usd" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase7-go-cam-causal-activity-modeling-2019", "title": "Gene Ontology Causal Activity Modeling (GO-CAM) moves beyond GO annotations to structured descriptions of biological functions and systems", "authors_or_org": "The Gene Ontology Consortium and collaborators", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/s41588-019-0500-1", "doi_or_identifier": "10.1038/s41588-019-0500-1", "venue_or_site": "Nature Genetics", "abstract_or_summary": "Introduces Gene Ontology Causal Activity Modeling, which moves beyond single GO annotations toward structured biological models of activities, causal relations, participants, and systems.", "key_claims": [ "GO-CAM represents biological function as structured causal activity models.", "GO-CAM links gene products, molecular activities, biological processes, and causal relations.", "Structured causal annotations make computational reasoning over biological systems more precise." ], "ontology_relevance": "Strong example of ontology moving from vocabulary to relation-rich executable scientific knowledge.", "ai_relevance": "Demonstrates how ontologies can structure causal machine-readable knowledge rather than only labels for model training or enrichment analysis.", "palantir_relevance": "Useful analogy for Palantir-style operational ontology: typed objects and relations become models of actions and causal processes.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biology", "causal-modeling", "gene-ontology", "go-cam", "knowledge-graph", "ontology-infrastructure", "phase7", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "sage-palantir-public-sector-2024", "title": "In Palantir we trust? Regulation of data analysis platforms in public administration", "authors_or_org": "SAGE journal article authors", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://journals.sagepub.com/doi/10.1177/20539517241255108", "doi_or_identifier": "10.1177/20539517241255108", "venue_or_site": "Big Data & Society", "abstract_or_summary": "Peer-reviewed article on regulation of data analysis platforms in public administration, using Palantir as a case to examine privacy, accountability, public-sector dependence, and platform governance.", "key_claims": [ "Public administrations using data analysis platforms face regulatory and accountability challenges.", "Palantir offers privacy and civil-liberties engineering, but platform use still depends on public-sector governance choices.", "The article highlights trust, regulation, and dependence questions.", "Public administration platforms raise issues of trust, opacity, accountability, and vendor dependence.", "Regulation must address platform operation, not only individual algorithms or datasets.", "The article is useful for framing public-sector governance concerns around Palantir deployments." ], "ontology_relevance": "Useful critique of ontology/data platforms as administrative infrastructure.", "ai_relevance": "Relevant to governance of AI/data-analysis platforms in public institutions.", "palantir_relevance": "Peer-reviewed source on Palantir and public administration regulation.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "academic", "governance", "palantir", "platform-governance", "privacy", "public-administration", "regulation", "trust", "vendor-dependence" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-star-griesemer-1989-boundary-objects", "title": "Institutional Ecology, 'Translations' and Boundary Objects: Amateurs and Professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39", "authors_or_org": "Susan Leigh Star and James R. Griesemer", "year": 1989, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1177/030631289019003001", "doi_or_identifier": "10.1177/030631289019003001", "venue_or_site": "Social Studies of Science", "abstract_or_summary": "Canonical STS paper introducing boundary objects as artifacts that are robust enough to coordinate across communities yet flexible enough for local use.", "key_claims": [ "Boundary objects enable cooperation without requiring full consensus.", "Classification systems, repositories, maps, and forms can coordinate heterogeneous social worlds.", "Standardization and local adaptation coexist in practical scientific infrastructure." ], "ontology_relevance": "Powerful frame for ontology as a coordination mechanism rather than a purely technical truth model.", "ai_relevance": "AI systems using enterprise ontologies must work across expert communities with different local meanings.", "palantir_relevance": "Useful for interpreting operational ontology as a boundary object among executives, operators, engineers, analysts, and AI agents.", "quality_signal": "widely_cited", "retrieval_tags": [ "boundary-objects", "classification", "coordination", "infrastructure", "phase3", "sts" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase2-graphrag-013", "title": "Knowledge Graph-Guided Retrieval Augmented Generation", "authors_or_org": "Xiangrong Zhu; Yuexiang Xie; Yi Liu; Yaliang Li; Wei Hu", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2502.06864", "doi_or_identifier": "arxiv:2502.06864", "venue_or_site": "arXiv", "abstract_or_summary": "Proposes KG-guided retrieval for multi-hop question answering, expanding seed chunks through knowledge graph relations to improve evidence coverage and organization.", "key_claims": [ "Knowledge graph expansion can retrieve coherent evidence chains beyond top-k vector chunks.", "Graph-guided organization helps assemble multi-hop context for generation.", "KG-RAG performance depends on entity linking and relation quality." ], "ontology_relevance": "Direct source for using an ontology or KG as a retrieval controller rather than a passive metadata layer.", "ai_relevance": "Recent KG-RAG method focused on multi-hop QA and graph-guided context assembly.", "palantir_relevance": "Relevant to traversing enterprise ontology links to gather evidence for agent answers.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "evidence-chain", "graph-guided-retrieval", "kg-rag", "kg2rag", "llm-grounding", "multi-hop-qa" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "p2-llm-ont-007", "title": "Large language models as oracles for instantiating ontologies with domain-specific knowledge", "authors_or_org": "Giovanni Ciatto; Andrea Agiollo; Matteo Magnini; Andrea Omicini", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2404.04108", "doi_or_identifier": "10.1016/j.knosys.2024.112940", "venue_or_site": "Knowledge-Based Systems", "abstract_or_summary": "Presents a domain-independent approach that queries LLMs with schema and query templates to instantiate ontologies with domain-specific entities and relations, evaluated on a nutrition case study.", "key_claims": [ "LLMs can populate ontology schemas faster when prompts are structured around classes, properties, and templates.", "Schema guidance reduces but does not eliminate erroneous entities and relations.", "Generated instances are best treated as candidates for expert keep, adjust, discard, or complement decisions." ], "ontology_relevance": "Strong peer-reviewed evidence for LLM-assisted ABox population under a predefined ontology schema.", "ai_relevance": "Shows how prompting can use symbolic schemas as constraints on generative model output.", "palantir_relevance": "Relevant to populating enterprise objects and relationships from domain descriptions while retaining expert governance.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "abox", "knowledge-based-systems", "llm", "ontology-population", "schema-guided-generation" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase6-lippolis-2025-llm-assisting-ontology-evaluation", "title": "Large Language Models Assisting Ontology Evaluation", "authors_or_org": "Anna Sofia Lippolis; Mohammad Javad Saeedizade; Robin Keskisarkka; Aldo Gangemi; Eva Blomqvist; Andrea Giovanni Nuzzolese", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/978-3-032-09527-5_27", "doi_or_identifier": "10.1007/978-3-032-09527-5_27", "venue_or_site": "ISWC 2025 / The Semantic Web", "abstract_or_summary": "Introduces OE-Assist for LLM-assisted competency-question verification and ontology evaluation in ontology-engineering workflows, with a controlled study involving ontology engineers.", "key_claims": [ "LLMs can assist competency-question verification at roughly human-user performance levels in the reported setting.", "Hybrid expert-plus-LLM evaluation can reduce ontology evaluation effort while retaining expert control.", "The paper contributes a dataset of competency questions paired with ontologies and ontology stories." ], "ontology_relevance": "High-value source for competency-question evaluation, ontology testing, and ontology quality workflows.", "ai_relevance": "Studies LLMs as ontology evaluators and workflow assistants, not only ontology generators.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "competency-questions", "evaluation", "iswc-2025", "llm", "ontology-engineering", "ontology-evaluation", "phase6", "validation" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase6-zhao-2025-llm-ontology-requirements-engineering", "title": "Leveraging Large Language Models for Ontology Requirements Engineering", "authors_or_org": "Yihang Zhao", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1007/978-3-031-99554-5_40", "doi_or_identifier": "10.1007/978-3-031-99554-5_40", "venue_or_site": "ESWC 2025 Satellite Events / LNCS", "abstract_or_summary": "Explores LLMs as knowledge-engineer assistants for eliciting better ontology requirements from domain experts, focusing on ontology requirements engineering, user stories, and competency questions.", "key_claims": [ "LLMs may reduce ambiguity and bias in collaborative ontology requirements elicitation.", "Ontology requirements are a promising intervention point before downstream ontology development.", "Requirements engineering connects user stories and competency questions to ontology scope." ], "ontology_relevance": "Targets the earliest stage of ontology engineering: requirements, user stories, and competency questions.", "ai_relevance": "Positions LLMs as dialogue and elicitation agents for domain experts during ontology design.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "competency-questions", "eswc-2025", "llm", "ontology-engineering", "ontology-requirements", "phase6", "user-stories" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "p2-llm-ont-015", "title": "OLaLa: Ontology Matching with Large Language Models", "authors_or_org": "Sven Hertling; Heiko Paulheim", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2311.03837", "doi_or_identifier": "10.1145/3587259.3627571", "venue_or_site": "K-CAP 2023 / arXiv", "abstract_or_summary": "Explores zero-shot and few-shot prompting with open LLMs for Ontology Alignment Evaluation Initiative tasks, including prompt design and use of examples.", "key_claims": [ "Few-shot LLM prompts can approach supervised ontology matching systems on some tasks.", "Prompt representation of graph and ontology context materially affects matching quality.", "LLM-based matching still needs candidate generation and validation choices around the model." ], "ontology_relevance": "Peer-reviewed bridge from classic ontology matching to LLM-assisted alignment.", "ai_relevance": "Shows prompt engineering as a practical substitute for large supervised alignment datasets in some settings.", "palantir_relevance": "Relevant to integrating heterogeneous enterprise schemas, acquired systems, and domain ontologies.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "alignment", "k-cap-2023", "llm", "oaei", "ontology-matching" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "ont-ai-034", "title": "On the Dangers of Stochastic Parrots: Can Language Models Be Too Big?", "authors_or_org": "Emily M. Bender; Timnit Gebru; Angelina McMillan-Major; Shmargaret Shmitchell", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1145/3442188.3445922", "doi_or_identifier": "10.1145/3442188.3445922", "venue_or_site": "FAccT 2021", "abstract_or_summary": "Critiques risks of large language models, including environmental cost, bias, documentation failures, and ungrounded text generation.", "key_claims": [ "Large language models can reproduce harmful patterns from training data.", "Fluent generation should not be equated with grounded understanding.", "Documentation, data governance, and evaluation are essential for responsible deployment." ], "ontology_relevance": "Highlights governance and grounding issues that ontologies can partly address but also inherit.", "ai_relevance": "Major responsible AI critique of LLM scaling.", "palantir_relevance": "Relevant to governance claims around enterprise AI, data lineage, and accountable use.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "governance", "grounding", "llm-risk", "responsible-ai" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-guarino-giaretta-1995-ontologies-kbs", "title": "Ontologies and Knowledge Bases: Towards a Terminological Clarification", "authors_or_org": "Nicola Guarino and Pierdaniele Giaretta", "year": 1995, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/FAIA950907", "doi_or_identifier": "10.3233/faia950907", "venue_or_site": "Towards Very Large Knowledge Bases, IOS Press", "abstract_or_summary": "Early formal-ontology paper clarifying competing meanings of ontology in AI, knowledge bases, and conceptual modeling.", "key_claims": [ "Ontology can mean a philosophical discipline, a conceptual system, or an engineered artifact depending on context.", "Confusion between concepts, languages, and specifications weakens knowledge engineering debates.", "Ontology work must distinguish intended domain commitments from implementation artifacts." ], "ontology_relevance": "Useful for article framing because it prevents conflating taxonomy, schema, graph database, and formal ontology.", "ai_relevance": "AI systems that expose symbolic interfaces need clarity about whether they use a vocabulary, a conceptual model, or a formal commitment layer.", "palantir_relevance": "Useful neutral lens for asking whether a product ontology is a formal semantic layer, an application object model, or both.", "quality_signal": "widely_cited", "retrieval_tags": [ "article-framing", "conceptual-modeling", "formal-ontology", "knowledge-bases", "phase3", "terminology" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "ont-ai-001", "title": "Ontology Learning for the Semantic Web", "authors_or_org": "Alexander Maedche; Steffen Staab", "year": 2001, "source_type": "paper", "bucket": "academic", "url": "https://ieeexplore.ieee.org/document/912103", "doi_or_identifier": "10.1109/5254.920602", "venue_or_site": "IEEE Intelligent Systems", "abstract_or_summary": "Early agenda-setting paper describing ontology learning as a way to semi-automatically acquire concepts, relations, and taxonomies for the Semantic Web.", "key_claims": [ "Ontology construction can be partially automated from text and data sources.", "Ontology learning requires multiple subtasks including term extraction, taxonomy induction, relation extraction, and evaluation.", "Human validation remains important for useful ontologies." ], "ontology_relevance": "Foundational framing for ontology learning as a pipeline rather than a single extraction task.", "ai_relevance": "Connects machine learning and NLP methods to symbolic knowledge acquisition.", "palantir_relevance": "Supports the idea that enterprise ontology maintenance can be assisted by AI but must remain governed.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "knowledge-acquisition", "ontology-learning", "semantic-web", "taxonomy-induction" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "euzenat_shvaiko_2013_ontology_matching", "title": "Ontology Matching, Second Edition", "authors_or_org": "Jerome Euzenat and Pavel Shvaiko", "year": 2013, "source_type": "book", "bucket": "books", "url": "https://doi.org/10.1007/978-3-642-38721-0", "doi_or_identifier": "10.1007/978-3-642-38721-0; isbn 9783642387203", "venue_or_site": "Springer", "abstract_or_summary": "Canonical book on finding correspondences between entities in different ontologies and using alignments to address semantic heterogeneity.", "key_claims": [ "Ontology matching addresses semantic heterogeneity among independently built ontologies.", "Correspondences can express equivalence, subsumption, disjointness, and other semantic relations.", "Matching systems require evaluation measures and benchmark datasets." ], "ontology_relevance": "Foundational source for ontology alignment and interoperability.", "ai_relevance": "Essential for AI systems integrating heterogeneous graphs, schemas, vocabularies, and domain models.", "palantir_relevance": "Directly relevant to enterprise integration where different source systems encode overlapping business concepts.", "quality_signal": "widely_cited", "retrieval_tags": [ "alignment", "interoperability", "ontology-matching", "semantic-heterogeneity" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase6-odps-41-open-data-product-specification", "title": "Open Data Product Specification 4.1", "authors_or_org": "Open Data Product Initiative / Linux Foundation", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://opendataproducts.org/v4.1", "doi_or_identifier": "odps 4.1", "venue_or_site": "Open Data Products", "abstract_or_summary": "Vendor-neutral, machine-readable metadata model for describing digital data products, including access, SLA, quality, licensing, pricing, and governance references.", "key_claims": [ "Data products need a shared, machine-readable structure.", "ODPS includes operational and business metadata beyond dataset schema.", "Version 4.1 adds reference mechanisms for modular governance." ], "ontology_relevance": "Useful source schema for modeling data products as first-class ontology entities.", "ai_relevance": "Lets agents reason about data product usability, access, quality, and commercial or policy constraints.", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "data-mesh", "data-product", "governance", "metadata-model", "odps", "phase6" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "industry-commercial-oracle-rdf-semantic-graph", "title": "Oracle RDF Semantic Graph documentation", "authors_or_org": "Oracle", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://docs.oracle.com/en/database/oracle/oracle-database", "doi_or_identifier": null, "venue_or_site": "Oracle Documentation", "abstract_or_summary": "Oracle documentation covers RDF graph storage, SPARQL querying, semantic indexing, and semantic technologies embedded in Oracle Database infrastructure.", "key_claims": [ "Enterprise databases can host RDF semantic graph workloads alongside traditional data management.", "Semantic graph capabilities can reuse database security, backup, and operational tooling.", "RDF and SPARQL support can integrate ontology workloads with existing database estates." ], "ontology_relevance": "Shows ontology infrastructure embedded in mainstream enterprise database platforms.", "ai_relevance": "Can provide governed semantic data to analytics and AI pipelines.", "palantir_relevance": "Contrasts proprietary operational ontology with database-native semantic graph support.", "quality_signal": "official_docs", "retrieval_tags": [ "enterprise-database", "ontology", "oracle", "rdf", "semantic-graph", "sparql" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-nissenbaum-2010-privacy-context", "title": "Privacy in Context: Technology, Policy, and the Integrity of Social Life", "authors_or_org": "Helen Nissenbaum", "year": 2010, "source_type": "book", "bucket": "books", "url": "https://www.sup.org/books/title?id=8862", "doi_or_identifier": "isbn 9780804752371", "venue_or_site": "Stanford University Press", "abstract_or_summary": "Major privacy theory book proposing contextual integrity: privacy depends on appropriate information flows within social contexts.", "key_claims": [ "Privacy is about appropriate flows of information, not only secrecy or individual control.", "Social contexts have norms governing actors, attributes, recipients, and transmission principles.", "Technical systems violate privacy when they disrupt contextual information norms." ], "ontology_relevance": "Critical lens for ontology/data integration: linking objects across contexts can violate contextual boundaries even when data is accurate.", "ai_relevance": "AI systems that retrieve and combine data need context-sensitive governance over information flows.", "palantir_relevance": "Highly relevant to public-sector and healthcare ontology platforms that integrate sensitive data across institutional contexts.", "quality_signal": "widely_cited", "retrieval_tags": [ "contextual-integrity", "critique", "data-governance", "information-flows", "phase3", "privacy" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase4-std-unesco-ai-ethics-recommendation-2021", "title": "Recommendation on the Ethics of Artificial Intelligence", "authors_or_org": "UNESCO", "year": 2021, "source_type": "policy_instrument", "bucket": "technical", "url": "https://unesdoc.unesco.org/ark:/48223/pf0000381137", "doi_or_identifier": null, "venue_or_site": "UNESCO Recommendation", "abstract_or_summary": "UNESCO recommendation establishing values, principles, and policy action areas for ethical AI, including human rights, environmental impacts, data governance, transparency, accountability, oversight, and international cooperation.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "accountability", "ai-ethics", "data-governance", "human-rights", "policy", "unesco" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-emery-trist-1960-sociotechnical-systems", "title": "Socio-Technical Systems", "authors_or_org": "F. E. Emery and E. L. Trist", "year": 1960, "source_type": "paper", "bucket": "academic", "url": "https://books.google.com/books?id=GoNQAAAAMAAJ", "doi_or_identifier": null, "venue_or_site": "Management Sciences, Models and Techniques", "abstract_or_summary": "Early formulation of socio-technical systems thinking in organizational design and management science.", "key_claims": [ "Organizations should be understood as jointly social and technical systems.", "The best technical design is not necessarily the best system design if social constraints are ignored.", "Joint optimization is needed across technology, tasks, people, and organization." ], "ontology_relevance": "Supports article framing that enterprise ontologies are coordination systems embedded in organizations.", "ai_relevance": "AI deployment should optimize model, data, workflow, human oversight, and institutional accountability together.", "palantir_relevance": "Useful for evaluating claims that ontology-based AI improves operations across complex organizations.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-deployment", "joint-optimization", "organization-design", "phase3", "sociotechnical-theory" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-bowker-star-1999-sorting-things-out", "title": "Sorting Things Out: Classification and Its Consequences", "authors_or_org": "Geoffrey C. Bowker and Susan Leigh Star", "year": 1999, "source_type": "book", "bucket": "books", "url": "https://mitpress.mit.edu/9780262522953/sorting-things-out", "doi_or_identifier": "isbn 9780262522953", "venue_or_site": "MIT Press", "abstract_or_summary": "Foundational book in infrastructure studies showing how classification systems organize work, produce visibility, and carry ethical and political consequences.", "key_claims": [ "Classifications are embedded in work practices, institutions, and infrastructures.", "Every classification system creates residual categories and consequences for those classified.", "Infrastructure becomes invisible when it works, but failures reveal social and political commitments." ], "ontology_relevance": "Central critique source: ontologies do not merely describe domains; they sort, prioritize, and exclude.", "ai_relevance": "AI systems trained or grounded on classifications inherit their categories, omissions, and institutional politics.", "palantir_relevance": "Important for analyzing enterprise ontology as a classification infrastructure with operational consequences.", "quality_signal": "widely_cited", "retrieval_tags": [ "classification", "critique", "ethics", "infrastructure-studies", "phase3", "residual-categories" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase6-idta-aas-metamodel-01001", "title": "Specification of the Asset Administration Shell Part 1: Metamodel, IDTA Number 01001", "authors_or_org": "Industrial Digital Twin Association", "year": 2024, "source_type": "docs", "bucket": "technical", "url": "https://industrialdigitaltwin.org/en?specificationpapers=specification-of-the-asset-administration-shell-part-1-metamodel-idta-number-01001", "doi_or_identifier": "idta 01001", "venue_or_site": "Industrial Digital Twin Association", "abstract_or_summary": "Defines the Asset Administration Shell metamodel for structured exchange of information about assets and Industry 4.0 components across value networks.", "key_claims": [ "AAS represents assets through a formal metamodel and submodels.", "The specification covers identifiers, exchange formats, access control, and mappings to XML, JSON, RDF, AutomationML, and OPC UA.", "The metamodel is intended for partner-to-partner lifecycle information exchange." ], "ontology_relevance": "High-value bridge between industrial digital twins and RDF/ontology representations.", "ai_relevance": "Gives agents a structured asset context model for industrial operations, maintenance, and decision support.", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "aas", "asset-administration-shell", "digital-twin", "industry-4-0", "metamodel", "phase6" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-lampland-star-2009-standards-stories", "title": "Standards and Their Stories: How Quantifying, Classifying, and Formalizing Practices Shape Everyday Life", "authors_or_org": "Martha Lampland and Susan Leigh Star, editors", "year": 2009, "source_type": "book", "bucket": "books", "url": "https://www.cornellpress.cornell.edu/book/9780801474610/standards-and-their-stories", "doi_or_identifier": "isbn 9780801474610", "venue_or_site": "Cornell University Press", "abstract_or_summary": "Edited volume on how standards, metrics, classifications, and formal systems reshape everyday work and institutional life.", "key_claims": [ "Standards are social and material arrangements, not neutral technical conveniences.", "Formalization changes the practices it claims to measure or coordinate.", "Standards distribute work, authority, and accountability across institutions." ], "ontology_relevance": "Provides critical vocabulary for ontology as standardization of organizational language and action.", "ai_relevance": "AI agents that depend on standards also amplify the institutional authority of those standards.", "palantir_relevance": "Useful for discussing ontology object/action types as operational standards within organizations.", "quality_signal": "widely_cited", "retrieval_tags": [ "classification", "critique", "formalization", "institutional-work", "phase3", "standards" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-busch-2011-standards-recipes-reality", "title": "Standards: Recipes for Reality", "authors_or_org": "Lawrence Busch", "year": 2011, "source_type": "book", "bucket": "books", "url": "https://mitpress.mit.edu/9780262518963/standards", "doi_or_identifier": "isbn 9780262518963", "venue_or_site": "MIT Press", "abstract_or_summary": "Book arguing that standards help produce the realities they appear merely to describe, structuring markets, organizations, and everyday life.", "key_claims": [ "Standards are recipes for ordering reality, not passive descriptions.", "The spread of standards can increase coordination while concentrating power in those who write and audit them.", "Technical standards have moral and political dimensions." ], "ontology_relevance": "Strong article frame for ontology as a standard that makes an operational reality actionable.", "ai_relevance": "AI systems can turn standards into automated decisions, raising stakes for who defines categories and thresholds.", "palantir_relevance": "Useful for analyzing ontology governance, vendor lock-in, and the authority of platform-defined categories.", "quality_signal": "widely_cited", "retrieval_tags": [ "classification", "governance", "markets", "phase3", "power", "standards" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-orlikowski-gash-1994-technological-frames", "title": "Technological Frames: Making Sense of Information Technology in Organizations", "authors_or_org": "Wanda J. Orlikowski and Debra C. Gash", "year": 1994, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/192844.193034", "doi_or_identifier": "10.1145/192844.193034", "venue_or_site": "ACM Transactions on Information Systems", "abstract_or_summary": "Paper introducing technological frames as the assumptions, expectations, and knowledge people use to understand technology in organizations.", "key_claims": [ "Different stakeholder groups interpret the same technology through different frames.", "Frame incongruence can cause implementation problems and organizational conflict.", "Successful technology change requires attention to meanings, expectations, and work practices." ], "ontology_relevance": "Explains why one shared ontology may be interpreted differently by departments, engineers, managers, and frontline users.", "ai_relevance": "AI governance needs to surface different stakeholder interpretations of model outputs, data categories, and tool permissions.", "palantir_relevance": "Useful for rollout and adoption analysis of enterprise ontology platforms across stakeholder groups.", "quality_signal": "widely_cited", "retrieval_tags": [ "adoption", "interpretation", "organizational-change", "phase3", "stakeholders", "technological-frames" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase3-orlikowski-1992-duality-technology", "title": "The Duality of Technology: Rethinking the Concept of Technology in Organizations", "authors_or_org": "Wanda J. Orlikowski", "year": 1992, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1287/orsc.3.3.398", "doi_or_identifier": "10.1287/orsc.3.3.398", "venue_or_site": "Organization Science", "abstract_or_summary": "Influential organizational theory paper arguing that technology is both shaped by and shapes human action in organizations.", "key_claims": [ "Technology is constituted through recurring human use while also structuring future action.", "Organizational technologies should be analyzed as enacted in practice, not fixed artifacts.", "Rules and resources embedded in technology affect power, coordination, and change." ], "ontology_relevance": "Useful for understanding ontology as enacted infrastructure that evolves through organizational use.", "ai_relevance": "AI systems and ontologies shape user behavior while being reshaped by feedback, edits, and workarounds.", "palantir_relevance": "Relevant to studying operational ontology adoption, user edits, and changing workflows over time.", "quality_signal": "widely_cited", "retrieval_tags": [ "enactment", "organization-theory", "phase3", "sociotechnical", "technology-in-practice" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "wilkinson_etal_2016_fair", "title": "The FAIR Guiding Principles for Scientific Data Management and Stewardship", "authors_or_org": "Mark D. Wilkinson, Michel Dumontier, IJsbrand Jan Aalbersberg, et al.", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/sdata.2016.18", "doi_or_identifier": "10.1038/sdata.2016.18", "venue_or_site": "Scientific Data", "abstract_or_summary": "Defines the FAIR principles: data and metadata should be findable, accessible, interoperable, and reusable by humans and machines.", "key_claims": [ "Machine-actionable metadata is central to reusable scientific data.", "Interoperability requires formal, shared, and broadly applicable languages for knowledge representation.", "Reusable data needs rich metadata, provenance, and clear usage conditions." ], "ontology_relevance": "Ontologies are a core mechanism for semantic interoperability and reusable metadata under FAIR.", "ai_relevance": "AI systems benefit from FAIR data because retrieval, provenance, integration, and reuse become more reliable.", "palantir_relevance": "Relevant to enterprise data products and ontology-backed governance of reusable operational data.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "data-stewardship", "fair", "interoperability", "metadata", "reuse" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase14-hoyt-gyori-2024-o3-guidelines-curated-resources", "title": "The O3 guidelines: open data, open code, and open infrastructure for sustainable curated scientific resources", "authors_or_org": "Charles Tapley Hoyt; Benjamin M. Gyori", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/s41597-024-03406-w", "doi_or_identifier": "10.1038/s41597-024-03406-w", "venue_or_site": "Scientific Data", "abstract_or_summary": "The O3 guidelines propose open data, open code, and open infrastructure practices for sustainable curated scientific resources. The Scientific Data article emphasizes technical workflows, social workflows, and progressive governance to keep community resources usable and maintainable.", "key_claims": [ "Curated scientific resources can become inaccessible or stale without open data, open code, open infrastructure, and sustainable governance.", "O3 combines technical workflows, social workflows, and progressive governance for community-facing curation.", "Ontology and knowledge-graph infrastructure should be evaluated as maintained resources, not only as one-time datasets." ], "ontology_relevance": "Supports the article's governance claim that ontology-backed AI depends on maintenance, contribution workflows, openness, and infrastructure sustainability.", "ai_relevance": "AI-ready ontologies and curated KGs are long-lived resources; O3 provides governance principles for sustainability, openness, automation, and community maintenance.", "palantir_relevance": "Useful contrast with proprietary ontology platforms: O3 frames sustainability and community governance as central to curated knowledge resources.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "community-curation", "knowledge-infrastructure", "o3-guidelines", "open-code", "open-data", "open-infrastructure", "phase14", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase11-tuan-sanyal-2026-predeployment-assurance-ontology-simulation", "title": "Toward Pre-Deployment Assurance for Enterprise AI Agents: Ontology-Grounded Simulation and Trust Certification", "authors_or_org": "Thanh Luong Tuan; Abhijit Sanyal", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.04037", "doi_or_identifier": "10.48550/arxiv.2606.04037", "venue_or_site": "arXiv", "abstract_or_summary": "Presents a pre-deployment verification framework for enterprise AI agents based on ontology-grounded simulation and machine-verifiable trust certification. It defines an Agent Operational Envelope, generates regulatory/operational/adversarial scenarios from ontologies, and evaluates scenarios across regulated industries before deployment.", "key_claims": [ "Pre-deployment verification of enterprise AI agents is a gap between LLM benchmarks and production deployment.", "An Agent Operational Envelope can formalize permissions, domain constraints, safety properties, governance rules, and autonomy levels.", "Ontology-to-scenario generation can derive regulatory, operational, and adversarial tests automatically.", "The framework proposes machine-verifiable Trust Certificates with graduated deployment verdicts." ], "ontology_relevance": "Important current source for ontology as assurance and scenario-generation infrastructure, not only knowledge representation.", "ai_relevance": "Frontier evidence for shifting agent assurance from post-deployment monitoring toward ontology-grounded pre-deployment scenario generation and certification.", "palantir_relevance": "Useful comparator for Palantir AIP evals, simulations, and governed deployments; treat as preprint evidence.", "quality_signal": "preprint_benchmark", "retrieval_tags": [ "agent-operational-envelope", "agentic-ai", "ai-assurance", "ontology-grounded-simulation", "phase11", "predeployment-assurance", "regulatory-coverage", "trust-certification", "validation" ], "triage_tier": "candidate", "triage_score": 54 }, { "id": "phase4-nature-huang-2025-llm-hallucination-survey", "title": "A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions", "authors_or_org": "Lei Huang, Weijiang Yu, Weitao Ma, Weihong Zhong, Zhangyin Feng, Haotian Wang, Qianglong Chen, Weihua Peng, Xiaocheng Feng, Bing Qin, Ting Liu", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2311.05232", "doi_or_identifier": "10.48550/arxiv.2311.05232", "venue_or_site": "ACM Transactions on Information Systems", "abstract_or_summary": "Focuses on hallucination in LLMs, including causes, detection, mitigation, retrieval limitations, and open problems around knowledge boundaries.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "hallucination", "llm", "rag", "reliability", "survey" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oa-https-openalex-org-w4393992111", "title": "A virtual reality-based approach for interactive and visual mining of association rules", "authors_or_org": "Zohra Ben Said", "year": 2012, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4393992111", "doi_or_identifier": null, "venue_or_site": "HAL (Le Centre pour la Communication Scientifique Directe)", "abstract_or_summary": "This thesis is at the intersection of two active esearch areas: Association Rules Mining and Virtual Reality. The main limitations of the association rule extraction algorithms are (i) the large amount of the generated rules and (ii) their low quality. Several solutions have been proposed to address this problem such as, the post-processing of association rules that allows rule validation and extraction of useful knowledge. Whereas rules are automatically extracted by combinatorial algorithms, rule post-processing is done by the user. Visualisation can help the user facing the large amount of rules by representing them in visual form. In order to find relevant knowledge in visual representations, the user needs to interact with these representations. To this aim, it is essential to provide the user with efficient interaction techniques. This work addresses two main issues: an association rule representation that allows the user quickly detection of the most interesting rules and interactive exploration of rules. The first issue requires an intuitive representation metaphor of association rules. The second requires an interactive exploration process allowing the user to explore the rule search space focusing on interesting rules. The main contributions of this work can be summarised as follows: (i) We propose a new classification for Visual Data Mining techniques, based on both 3D representations and interaction techniques. Such a classification helps the user choosing a visual representation and an interaction technique for his/her application. (ii) We propose a new visualisation metaphor for association rules that takes into account the attributes of the rule, the contribution of each one, and their correlations. (iii) We propose a methodology for interactive exploration of associationrules to facilitate the user task facing large sets of rules taking into account his/her cognitive capabilities. In this methodology, local algorithms are used to recommend better rules based on a reference rule which is proposed by the user. Then, the user can both drives extraction and post-processing of rules using appropriate interaction operators. (iv) We developed a tool that implements all the methodology functionality. The tool is based on an intuitive display in a virtual environment and supports multiple interaction methods.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oa-https-doi-org-10-1016-j-polgeo-2024-103134", "title": "A world model: On the political logics of generative AI", "authors_or_org": "Louise Amoore, Alexander Campolo, Benjamin N. Jacobsen, Ludovico Rella", "year": 2024, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4398792098", "doi_or_identifier": "10.1016/j.polgeo.2024.103134", "venue_or_site": "Political Geography", "abstract_or_summary": "The computational logics of large language models (LLMs) or generative AI – from the early models of CLIP and BERT to the explosion of text and image generation via ChatGPT and DALL-E − are increasingly penetrating the social and political world. Not merely in the direct sense that generative AI models are being deployed to govern difficult problems, whether decisions on the battlefield or responses to pandemic, but also because generative AI is shaping and delimiting the political parameters of what can be known and actioned in the world. Contra the promise of a generalizable “world model” in computer science, the article addresses how and why generative AI gives rise to a model of the world, and with it a set of political logics and governing rationalities that have profound and enduring effects on how we live today. The article traces the genealogies of generative AI models, how they have come into being, and why some concepts and techniques that animate these models become durable forms of knowledge that actively shape the world, even long after a specific material commercial GPT model has moved on to a new iteration. Though generative AI retains significant traces of former scientific and computational regimes – in statistical practices, probabilistic knowledge, and so on – it is also dislocating epistemological arrangements and opening them to novel ways of perceiving, characterising, classifying, and knowing the world. Four defining aspects of the political logic of generative AI are elaborated: i) generativity as something more than the capacity to generate image or text outputs, so that a generative logic acts upon the world understood as estimates of “underlying distributions” in data; ii) latency as a political logic of compression in which (by contrast with claims to reduction or distortion) the thing that is hidden, unknown or latent becomes surfaced and amenable to being governed; iii) broken and parallelized sequences as the ordering device of the political logic of generative AI, where attention frameworks radically change the possibilities for governing non-linear problems; iv) pre-training and fine-tuning as a computational logic of generative AI that simultaneously shapes a “zero shot politics” oriented towards unencountered data and new tasks. Across each of the four aspects, the article maps the emerging contemporary political logic of generative AI.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase2-wef-ai-agents-governance-2025", "title": "AI Agents in Action: Foundations for Evaluation and Governance", "authors_or_org": "World Economic Forum", "year": 2025, "source_type": "technical_report", "bucket": "technical", "url": "https://www.weforum.org/publications/ai-agents-in-action", "doi_or_identifier": null, "venue_or_site": "World Economic Forum", "abstract_or_summary": "World Economic Forum report on evaluating and governing AI agents, including autonomy, tool use, risk tiers, oversight, evaluation, and accountability patterns.", "key_claims": [ "Agent governance must account for tool use, autonomy, environment access, and downstream consequences.", "Evaluation should cover plans, tool calls, task completion, safety, reliability, and accountability, not only model output quality.", "Ontology-exposed tools and action types make agent governance more concrete but also raise the stakes for authorization and audit." ], "ontology_relevance": "Connects ontology-as-tool-interface to practical governance requirements for AI agents.", "ai_relevance": "Useful synthesis source for agent evaluation and governance patterns.", "palantir_relevance": "Directly relevant to Palantir Ontology MCP and AIP agent workflows that expose read/write tools.", "quality_signal": "secondary_source", "retrieval_tags": [ "accountability", "agents", "autonomy", "evaluation", "governance", "mcp", "tool-use" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-omg-api4kp-1-0-2023", "title": "API4KP: Application Programming Interfaces for Knowledge Platforms, Version 1.0", "authors_or_org": "Object Management Group", "year": 2023, "source_type": "standard", "bucket": "technical", "url": "https://www.omg.org/spec/API4KP/1.0", "doi_or_identifier": null, "venue_or_site": "OMG Formal Specification", "abstract_or_summary": "OMG specification for APIs and service abstractions supporting knowledge assets, knowledge processing, knowledge representation, and platform interoperability.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "api", "api4kp", "governance", "knowledge-asset", "knowledge-platform", "omg", "semantic-interoperability" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "s2-10-3389-fdgth-2025-1552372", "title": "Artificial intelligence in nursing: an integrative review of clinical and operational impacts", "authors_or_org": "Salwa Hassanein, R. E. Arab, A. Abdrbo, Mohammad S. Abu-Mahfouz, Mastoura Khames Farag Gaballah, M. Seweid, Mohammed Almari, Husam Alzghoul", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/e954dcd7fbea349105d21de0257dad282d62c03b", "doi_or_identifier": "10.3389/fdgth.2025.1552372", "venue_or_site": "Frontiers Digit. Health", "abstract_or_summary": "Background Advances in digital technologies and artificial intelligence (AI) are reshaping healthcare delivery, with AI increasingly integrated into nursing practice. These innovations promise enhanced diagnostic precision, improved operational workflows, and more personalized patient care. However, the direct impact of AI on clinical outcomes, workflow efficiency, and nursing staff well-being requires further elucidation. Methods This integrative review synthesized findings from 18 studies published through November 2024 across diverse healthcare settings. Using the PRISMA 2020 and SPIDER frameworks alongside rigorous quality appraisal tools (MMAT and ROBINS-I), the review examined the multifaceted effects of AI integration in nursing. Our analysis focused on three principal domains: clinical advancements and patient monitoring, operational efficiency and workload management, and ethical implications. Results The review demonstrates that AI integration in nursing has yielded substantial clinical and operational benefits. AI-powered monitoring systems, including wearable sensors and real-time alert platforms, have enabled nurses to detect subtle physiological changes—such as early fever onset or pain indicators—well before traditional methods, resulting in timely interventions that reduce complications, shorten hospital stays, and lower readmission rates. For example, several studies reported that early-warning algorithms facilitated faster clinical responses, thereby improving patient safety and outcomes. Operationally, AI-based automation of routine tasks (e.g., scheduling, administrative documentation, and predictive workload classification) has streamlined resource allocation. These efficiencies have led to a measurable reduction in nurse burnout and improved job satisfaction, as nurses can devote more time to direct patient care. However, despite these benefits, ethical challenges remain prominent. Key concerns include data privacy risks, algorithmic bias, and the potential erosion of clinical judgment due to overreliance on technology. These issues underscore the need for robust ethical frameworks and targeted AI literacy training within nursing curricula. Conclusion This review demonstrates that AI integration holds transformative potential for nursing practice by enhancing both clinical outcomes and operational efficiency. However, to realize these benefits fully, it is imperative to develop robust ethical frameworks, incorporate comprehensive AI literacy training into nursing education, and foster interdisciplinary collaboration. Future longitudinal studies across varied clinical contexts are essential to validate these findings and support the sustainable, equitable implementation of AI technologies in nursing. Policymakers and healthcare leaders must prioritize investments in AI solutions that complement the expertise of nursing professionals while addressing ethical risks.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oa-https-doi-org-10-12688-openreseurope-16536-1", "title": "Artificial Intelligence Technologies and Practical Normativity/Normality: Investigating Practices beyond the Public Space", "authors_or_org": "Ingvild Bode, Hendrik Huelss", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4392649576", "doi_or_identifier": "10.12688/openreseurope.16536.2", "venue_or_site": "Open Research Europe", "abstract_or_summary": "This essay examines how artificial intelligence (AI) technologies may shape international norms. Following a brief discussion of the ways in which AI technologies pose new governance questions, we reflect on the extent to which norm research in the discipline of International Relations (IR) is equipped to understand how AI technologies shape normative substance. Norm research has typically focused on the impact and failure of norms, offering increasingly diversified models of norm contestation, for instance. But present research has two shortcomings: a near-exclusive focus on modes and contexts of norm emergence and constitution that happen in the public space; and a focus on the workings of a pre-set normativity (ideas of oughtness and justice) that stands in an unclear relationship with normality (ideas of the standard, the average) emerging from practices. Responding to this, we put forward a research programme on AI and practical normativity/normality based on two pillars: first, we argue that operational practices of designing and using AI technologies typically performed outside of the public eye make norms; and second, we emphasise the interplay of normality and normativity as analytically influential in this process. With this, we also reflect on how increasingly relying on AI technologies across diverse policy domains has an under-examined effect on the exercise of human agency. This is important because the normality shaped by AI technologies can lead to forms of non-human generated normativity that risks replacing conventional models about how norms matter in AI-affected policy domains. We conclude that AI technologies are a major, yet still under-researched, challenge for understanding and studying norms. We should therefore reflect on new theoretical perspectives leading to insights that are also relevant for the struggle about top-down forms of AI regulation.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase3-llm-ontmem-027", "title": "Assessing the Effectiveness of Ontology-Grounded AI Term Extraction", "authors_or_org": "Ontology-grounded term extraction authors", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC12892472", "doi_or_identifier": null, "venue_or_site": "PMC / biomedical informatics article", "abstract_or_summary": "Evaluates ontology-grounded AI term extraction, including OntoGPT-style grounding processes that map extracted text spans and concepts to ontology terms.", "key_claims": [ "Ontology grounding changes information extraction from free text labeling into normalized semantic curation.", "Term extraction should be evaluated for both extraction accuracy and correctness of ontology identifier grounding.", "Biomedical use cases reveal the importance of curated vocabularies for reducing ambiguity in LLM outputs." ], "ontology_relevance": "Adds evaluation evidence for ontology-grounded extraction and term normalization.", "ai_relevance": "Shows how LLM extraction can be constrained and evaluated through curated biomedical ontologies.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "biomedical-ai", "identifier-normalization", "ontogpt", "ontology-grounding", "term-extraction" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-omg-col-1-0-2023", "title": "Commons Ontology Library, Version 1.0", "authors_or_org": "Object Management Group", "year": 2023, "source_type": "standard", "bucket": "technical", "url": "https://www.omg.org/spec/COL/1.0", "doi_or_identifier": null, "venue_or_site": "OMG Formal Specification", "abstract_or_summary": "Defines reusable ontology modules for common concepts intended to support semantic interoperability across OMG and enterprise modeling specifications.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "commons-ontology-library", "enterprise-modeling", "omg", "ontology-modules", "semantic-interoperability" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "p2-llm-ont-032", "title": "Data Catalog Vocabulary (DCAT) - Version 3", "authors_or_org": "World Wide Web Consortium", "year": 2024, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/vocab-dcat-3", "doi_or_identifier": "w3c:rec-vocab-dcat-3-20240822", "venue_or_site": "W3C Recommendation", "abstract_or_summary": "W3C Recommendation defining DCAT Version 3 for describing catalogs, datasets, data services, distributions, and related metadata on the Web.", "key_claims": [ "Standard catalog metadata improves discoverability and interoperability of datasets and data services.", "DCAT can act as a metadata layer for decentralized data publishing and federated discovery.", "Catalog semantics are a practical bridge between data management and ontology-backed applications." ], "ontology_relevance": "Core standard for semantic cataloging and enterprise data product discovery.", "ai_relevance": "Provides structured metadata that can ground LLM retrieval, data discovery, and tool selection.", "palantir_relevance": "Useful standards comparator for ontology-backed catalog and data-product layers.", "quality_signal": "official_standard", "retrieval_tags": [ "data-catalog", "data-products", "dcat", "metadata", "rdf", "semantic-layer", "w3c" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-w3c-dqv-2016", "title": "Data on the Web Best Practices: Data Quality Vocabulary", "authors_or_org": "Riccardo Albertoni, Antoine Isaac, W3C Data on the Web Best Practices Working Group", "year": 2016, "source_type": "standard_note", "bucket": "technical", "url": "https://www.w3.org/TR/vocab-dqv", "doi_or_identifier": null, "venue_or_site": "W3C Working Group Note", "abstract_or_summary": "W3C vocabulary for describing dataset quality measurements, quality dimensions, quality annotations, certificates, conformance, policies, and provenance-linked quality metadata.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "data-quality", "dcat", "governance", "metadata", "provenance", "semantic-interoperability", "w3c" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase3-llm-ontmem-030", "title": "Data-Centric Artificial Intelligence: A Survey", "authors_or_org": "Daochen Zha; Zaid Pervaiz Bhat; Kwei-Herng Lai; Fan Yang; Xia Hu", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2303.10158", "doi_or_identifier": "arxiv:2303.10158", "venue_or_site": "ACM Computing Surveys / arXiv", "abstract_or_summary": "Survey of data-centric AI covering data quality, labeling, cleaning, augmentation, governance, and lifecycle management as first-class drivers of AI performance.", "key_claims": [ "AI performance depends heavily on data quality and data management, not only model architecture.", "Data-centric workflows require lifecycle governance, evaluation, and iterative improvement of data assets.", "Ontology and semantic-layer work can be framed as part of data-centric AI infrastructure for reliable models." ], "ontology_relevance": "Provides broader data-centric AI context for treating ontology and semantic layers as production AI infrastructure.", "ai_relevance": "Supports the argument that high-value AI depends on maintained external data and metadata assets.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "ai-infrastructure", "data-centric-ai", "data-quality", "governance", "semantic-layer" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-w3c-did-core-2022", "title": "Decentralized Identifiers (DIDs) v1.0", "authors_or_org": "Manu Sporny, Amy Guy, Markus Sabadello, Drummond Reed, W3C Decentralized Identifier Working Group", "year": 2022, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/did-1.0", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines decentralized identifiers, DID documents, verification methods, services, and DID resolution concepts for controllable, verifiable identifiers.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "did", "governance", "identity", "provenance", "trust", "verifiable-data", "w3c" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase6-omg-dmn-15", "title": "Decision Model and Notation Specification Version 1.5", "authors_or_org": "Object Management Group", "year": 2024, "source_type": "docs", "bucket": "technical", "url": "https://www.omg.org/spec/DMN/1.5/About-DMN", "doi_or_identifier": "omg dmn 1.5", "venue_or_site": "Object Management Group", "abstract_or_summary": "Standard notation and expression language for modeling operational decisions, decision requirements, business knowledge, input data, and decision tables.", "key_claims": [ "DMN bridges business decision design and technical implementation.", "Decision models can be shared between business analysts, developers, and decision owners.", "DMN is designed to complement BPMN process modeling." ], "ontology_relevance": "Useful for representing decision logic linked to ontology concepts, policies, and business rules.", "ai_relevance": "Can constrain or explain agent recommendations through explicit decision models and rules.", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "business-rules", "decision-intelligence", "decision-model", "dmn", "feel", "omg", "phase6" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-omg-dmn-1-6-2024", "title": "Decision Model and Notation, Version 1.6", "authors_or_org": "Object Management Group", "year": 2024, "source_type": "standard", "bucket": "technical", "url": "https://www.omg.org/spec/DMN/1.6", "doi_or_identifier": null, "venue_or_site": "OMG Formal Specification", "abstract_or_summary": "Defines notation, metamodel, and expression language for decision requirements, decision logic, decision tables, business knowledge models, and input data.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "decision-modeling", "dmn", "explainability", "governance", "omg", "policy-automation", "rules" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase3-kent-nhs-fdp-privacy-notice-2026", "title": "Federated Data Platform (FDP) product privacy notice", "authors_or_org": "Kent Community Health NHS Foundation Trust", "year": 2026, "source_type": "webpage", "bucket": "synthesis", "url": "https://www.kentcht.nhs.uk/privacy/project-and-service-specific-privacy-notices/palantir-federated-data-platform-privacy-notice", "doi_or_identifier": null, "venue_or_site": "Kent Community Health NHS Foundation Trust", "abstract_or_summary": "Local NHS privacy notice explaining access controls, approved purposes, confidentiality, governance, and Palantir's processor role under NHS instruction for an FDP product.", "key_claims": [ "Access to NHS health and social care data within FDP is described as carefully controlled.", "Only authorized users are granted access for approved purposes.", "Palantir is described as operating under NHS instruction as a processor." ], "ontology_relevance": "Shows local privacy framing around platform-mediated operational data use.", "ai_relevance": "Relevant to data access boundaries for AI or analytics built on FDP data.", "palantir_relevance": "Public-sector privacy evidence for Palantir processor-position claims.", "quality_signal": "primary_source", "retrieval_tags": [ "access-control", "fdp", "nhs", "palantir", "privacy-notice", "processor" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "s2-10-1109-access-2025-3530683", "title": "From ChatGPT to Sora: Analyzing Public Opinions and Attitudes on Generative Artificial Intelligence in Social Media", "authors_or_org": "Wenyan Feng, Yuhang Li, Chunhao Ma, Lisai Yu", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/f51d5ef166aa2f0773f4baafc736cd0f21c7146c", "doi_or_identifier": "10.1109/access.2025.3530683", "venue_or_site": "IEEE Access", "abstract_or_summary": "This study examines public opinions, emotional tendencies, and psychological linguistic characteristics associated with the launch of OpenAI’s ChatGPT and the advanced video generation model, Sora, by analyzing discussions on the Chinese social media platform Weibo. A total of 24,727 valid user-generated texts (1,762,296 words) were collected and analyzed using Python and its associated APIs. Word co-occurrence network analysis, topic modeling based on Latent Dirichlet Allocation (LDA), and emotional characteristics based on the DLUT Emotion Ontology and psycholinguistic analyses based on the Linguistic Inquiry and Word Count (LIWC) dictionary were employed to explore public views on these generative AI technologies. The findings reveal a shift in public focus over time, from initial excitement about technological advancements to growing interest in commercialization, labor, education, ethics, and global competition. The public’s emotional responses to AI were a mix of excitement and apprehension. The study identifies seven distinct emotional types, providing a nuanced understanding of public psychological reactions, which contrasts with previous binary classifications. This research contributes valuable insights for policymakers, businesses, and researchers, highlighting the public’s evolving acceptance of generative AI technologies.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "industry-commercial-gartner-knowledge-graphs-data-fabric", "title": "Gartner-style narratives on knowledge graphs, data fabric, and generative AI", "authors_or_org": "Gartner", "year": 2024, "source_type": "commercial_article", "bucket": "commercial", "url": "https://www.gartner.com/en/information-technology/topics/data-fabric", "doi_or_identifier": null, "venue_or_site": "Gartner", "abstract_or_summary": "Gartner data fabric and AI narratives often place knowledge graphs and active metadata at the center of semantic integration, governance, and AI-ready data architecture.", "key_claims": [ "Knowledge graphs are marketed as a semantic foundation for data fabric and active metadata.", "Analyst language influences enterprise buyers even when detail is abstract.", "The narrative links KG adoption to data integration, governance, decision intelligence, and generative AI readiness." ], "ontology_relevance": "Important market framing for ontology as enterprise semantic infrastructure.", "ai_relevance": "Explains why knowledge graphs are increasingly tied to generative AI governance and grounding.", "palantir_relevance": "Useful for comparing Palantir's ontology narrative with broader enterprise analyst categories.", "quality_signal": "secondary_source", "retrieval_tags": [ "active-metadata", "data-fabric", "gartner", "generative-ai", "governance", "knowledge-graph", "market-framing", "market-narrative" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase3-llm-ontmem-018", "title": "Graph-based Agent Memory: Taxonomy, Techniques, and Evaluation", "authors_or_org": "Graph-based Agent Memory survey authors", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2602.05665", "doi_or_identifier": "10.48550/arxiv.2602.05665", "venue_or_site": "arXiv", "abstract_or_summary": "Survey of graph-based memory for LLM agents, organizing techniques, memory structures, retrieval mechanisms, multi-agent sharing, and evaluation challenges.", "key_claims": [ "Graph memory is emerging as a distinct agent-memory family rather than a minor variant of vector retrieval.", "Agent memory evaluation must include update, retrieval, temporal reasoning, access control, and downstream task success.", "Graph memory systems inherit ontology/KG problems such as entity resolution, stale edges, provenance, and schema drift.", "Graphs are a natural substrate for relational, hierarchical, and evolving agent memory.", "Agent memory should be analyzed across extraction, storage, retrieval, and evolution lifecycle phases.", "Benchmarks and open resources for graph-based memory remain fragmented." ], "ontology_relevance": "Useful synthesis source connecting ontology/KG engineering concerns to agent memory design.", "ai_relevance": "Provides taxonomy for graph-structured memory architectures in LLM agents.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "agent-memory", "arxiv-2026", "evaluation", "graph-based-memory", "graph-memory", "llm-agents", "memory-evolution", "memory-retrieval", "phase6", "survey", "temporal-reasoning" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-nature-noy-2019-industry-scale-kgs", "title": "Industry-Scale Knowledge Graphs: Lessons and Challenges", "authors_or_org": "Natasha Noy, Yuqing Gao, Anshu Jain, Anant Narayanan, Alan Patterson, Jamie Taylor", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3331166", "doi_or_identifier": "10.1145/3331166", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "Synthesizes lessons from large-scale industrial knowledge graphs, emphasizing schema evolution, entity resolution, quality, and organizational workflows.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "data-integration", "entity-resolution", "industrial-ai", "knowledge-graph", "schema-evolution" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase2-anthropic-mcp-2024", "title": "Introducing the Model Context Protocol", "authors_or_org": "Anthropic", "year": 2024, "source_type": "technical_article", "bucket": "technical", "url": "https://www.anthropic.com/news/model-context-protocol", "doi_or_identifier": null, "venue_or_site": "Anthropic News", "abstract_or_summary": "Announcement and technical framing of MCP as an open standard for secure two-way connections between data sources and AI-powered tools.", "key_claims": [ "MCP is positioned as a way to replace fragmented custom integrations with a reusable protocol.", "The architecture separates MCP clients inside AI applications from MCP servers exposing data and tools.", "The protocol is central to the emerging agent interoperability stack." ], "ontology_relevance": "Useful for discussing how ontology resources become agent-accessible interfaces.", "ai_relevance": "Primary source for the origin and motivation of MCP in agentic AI systems.", "palantir_relevance": "Provides non-Palantir baseline for understanding Palantir MCP and Ontology MCP.", "quality_signal": "primary_source", "retrieval_tags": [ "agents", "anthropic", "context", "interoperability", "mcp", "tool-use" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-iso-iec-5259-5-2025", "title": "ISO/IEC 5259-5:2025 Artificial intelligence - Data quality for analytics and machine learning (ML) - Part 5: Data quality governance framework", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2025, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/84150.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for governance-level oversight and direction of data quality for analytics and machine learning across organizations and data life cycles.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "accountability", "board-oversight", "data-quality-governance", "iec", "iso", "lifecycle", "machine-learning" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "nhs-fdp-privacy-policy-2026", "title": "NHS Federated Data Platform privacy policy", "authors_or_org": "NHS England", "year": 2026, "source_type": "webpage", "bucket": "technical", "url": "https://www.england.nhs.uk/digitaltechnology/nhs-federated-data-platform/privacy-policy", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "NHS England privacy-policy page for the Federated Data Platform, relevant to patient-facing explanations of data handling and rights.", "key_claims": [ "The privacy policy provides patient-facing information about FDP data handling.", "It is a primary source for stated privacy protections and public communication.", "The content should be compared with external critiques about transparency and trust." ], "ontology_relevance": "Useful for studying how public institutions explain operational data integration to affected populations.", "ai_relevance": "Foundational governance context for AI-ready healthcare data infrastructure.", "palantir_relevance": "Official public-sector source in a Palantir-associated deployment.", "quality_signal": "primary_source", "retrieval_tags": [ "fdp", "governance", "nhs", "palantir", "patients", "privacy-policy" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "nhs-fdp-contract-explainer-2026", "title": "NHS Federated Data Platform: Contract explainer", "authors_or_org": "NHS England", "year": 2026, "source_type": "webpage", "bucket": "technical", "url": "https://www.england.nhs.uk/digitaltechnology/nhs-federated-data-platform/security-privacy/contract-explainer", "doi_or_identifier": null, "venue_or_site": "NHS England", "abstract_or_summary": "Official NHS England explainer for FDP contract arrangements, supplier responsibilities, data-controller relationships, and security/privacy commitments.", "key_claims": [ "The contract explainer sets out NHS England's public account of supplier responsibilities and controls.", "It is useful for checking what Palantir is contractually claimed to do and not do.", "The page provides official language for governance, privacy, and public accountability claims.", "A Palantir-led consortium was awarded the NHS FDP contract in November 2023.", "The contract covers a seven-year period and can support up to 240 NHS organizations.", "The contract is publicly available through UK contract-publication channels." ], "ontology_relevance": "Shows how operational data-platform governance is described for a national public-sector deployment.", "ai_relevance": "Relevant to governing AI-ready data infrastructure and vendor access to sensitive data.", "palantir_relevance": "Primary public-sector source on Palantir-linked NHS FDP contract governance.", "quality_signal": "primary_source", "retrieval_tags": [ "contract", "fdp", "federated-data-platform", "governance", "health-data", "nhs", "palantir", "privacy", "public-sector", "security" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oaei_official", "title": "Ontology Alignment Evaluation Initiative", "authors_or_org": "OAEI organizers", "year": 2004, "source_type": "webpage", "bucket": "academic", "url": "https://oaei.ontologymatching.org", "doi_or_identifier": "oaei benchmark initiative", "venue_or_site": "Ontology Alignment Evaluation Initiative", "abstract_or_summary": "Community evaluation campaign and benchmark suite for comparing ontology matching systems across tracks and years.", "key_claims": [ "Ontology matching systems need shared benchmarks and reference alignments.", "Different matching tasks stress different methods and assumptions.", "Evaluation culture is necessary for progress in ontology alignment.", "Ontology alignment requires shared benchmarks and comparable evaluation protocols.", "Different matching tasks test different aspects of schema and instance alignment.", "Alignment quality is central to semantic interoperability." ], "ontology_relevance": "Primary benchmark ecosystem for ontology alignment research.", "ai_relevance": "Useful for evaluating LLM-assisted or embedding-assisted ontology mapping tools against established tasks.", "palantir_relevance": "Relevant to measuring semantic reconciliation across enterprise systems rather than relying on ad hoc mappings.", "quality_signal": "primary_source", "retrieval_tags": [ "benchmark", "evaluation", "matching", "oaei", "ontology-alignment", "schema-matching" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase3-llm-ontmem-028", "title": "Ontology Learning from Text: An Analysis on LLM Performance", "authors_or_org": "Ontology learning from text authors", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://ceur-ws.org/Vol-3874/paper5.pdf", "doi_or_identifier": "ceur-ws vol-3874 paper5", "venue_or_site": "CEUR-WS", "abstract_or_summary": "Analyzes LLM performance on ontology learning from text, including ontology conformance and hallucination categories for subject, relation, and object extraction in a safety-domain use case.", "key_claims": [ "Ontology learning evaluation should include ontology conformance, not only precision, recall, and F1.", "Subject, relation, and object hallucinations are distinguishable failure modes in LLM-extracted ontology content.", "Safety-domain ontology extraction illustrates the need to integrate multiple information sources and validation criteria." ], "ontology_relevance": "Adds task-level evaluation categories for LLM ontology extraction from text.", "ai_relevance": "Useful for diagnosing hallucination in structured extraction outputs.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "hallucination", "ontology-conformance", "ontology-learning", "safety-domain", "text-extraction" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase16-opc-ua-address-space-model-10000-3", "title": "OPC Unified Architecture Part 3: Address Space Model", "authors_or_org": "OPC Foundation", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://reference.opcfoundation.org/Core/Part3", "doi_or_identifier": "opc 10000-3", "venue_or_site": "OPC Foundation Specification", "abstract_or_summary": "OPC UA Part 3 defines the Address Space Model, the information-model foundation of OPC UA. It specifies nodes, references, attributes, types, objects, variables, methods, views, and event structures for industrial interoperability.", "key_claims": [ "OPC UA uses an address-space information model of nodes, references, attributes, types, objects, variables, methods, and events.", "Industrial interoperability depends on shared machine-readable information models, not only transport protocols.", "Operational AI in manufacturing can use information models as asset and action semantics." ], "ontology_relevance": "High-value industrial semantic standard for information modeling, digital twins, and operational objects.", "ai_relevance": "Industrial AI and digital-twin agents need machine-readable operational asset models, type systems, and methods that are separate from free-form prompts.", "palantir_relevance": "Comparator for Palantir operational objects and action semantics in industrial settings.", "quality_signal": "official_standard", "retrieval_tags": [ "address-space-model", "digital-twin", "industrial-semantic-standards", "industry-4-0", "information-model", "opc-ua", "operational-objects", "phase16" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-nature-w3c-owl-2004-overview", "title": "OWL Web Ontology Language Overview", "authors_or_org": "Deborah L. McGuinness, Frank van Harmelen", "year": 2004, "source_type": "standard", "bucket": "academic", "url": "https://www.w3.org/TR/owl-features", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Introduces OWL as a web ontology language for defining classes, properties, individuals, and logical constraints over web data.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "ontology-language", "owl", "semantic-web", "standards" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase2-graphrag-020", "title": "RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval", "authors_or_org": "Parth Sarthi; Salman Abdullah; Aditi Tuli; Shubh Khanna; Anna Goldie; Christopher D. Manning", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2401.18059", "doi_or_identifier": "arxiv:2401.18059", "venue_or_site": "ICLR 2024 / arXiv", "abstract_or_summary": "Introduces a hierarchical retrieval method that recursively clusters and summarizes text chunks into a tree for multi-level retrieval.", "key_claims": [ "Hierarchical summaries improve retrieval for questions requiring broad context.", "Higher-level abstractions can complement low-level chunks in RAG.", "Tree organization is an alternative to graph communities for corpus-level summarization." ], "ontology_relevance": "Useful comparator: ontologies and GraphRAG provide typed graph hierarchies, while RAPTOR provides unsupervised summary trees.", "ai_relevance": "Important advanced RAG baseline for hierarchical context retrieval.", "palantir_relevance": "Relevant to summarizing enterprise corpora when formal object graphs are unavailable or incomplete.", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "abstractive-summarization", "hierarchical-retrieval", "rag", "raptor", "summary-tree" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase6-parliament-rewiring-state-2026", "title": "Rewiring the State: Digital Centre of Government", "authors_or_org": "House of Commons Science, Innovation and Technology Committee", "year": 2026, "source_type": "webpage", "bucket": "technical", "url": "https://committees.parliament.uk/publications/53352/documents/298462/default", "doi_or_identifier": "hc committee report, published 2026-06-03", "venue_or_site": "UK Parliament Committees", "abstract_or_summary": "Cross-party committee report on digital government that singles out Palantir as a concerning example of UK public-sector dependence on a small group of major technology providers.", "key_claims": [ "The report argues Palantir should not have such a significant UK public-sector role.", "It calls Palantir's public-sector presence an unacceptable point of weakness.", "It recommends exercising the February 2027 FDP break clause.", "It recommends publishing a costed FDP exit plan by the end of 2026." ], "ontology_relevance": "Frames middleware and data-analysis platforms as infrastructure, which is central to ontology lock-in concerns.", "ai_relevance": "Connects public-sector AI and digital-state ambitions to vendor lock-in, sovereignty, and data hygiene risks.", "palantir_relevance": "High-authority public-sector scrutiny of Palantir's UK footprint.", "quality_signal": "primary_source", "retrieval_tags": [ "digital-sovereignty", "fdp", "lock-in", "nhs", "palantir", "parliament", "phase6", "public-sector" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-w3c-sosa-ssn-2017", "title": "Semantic Sensor Network Ontology", "authors_or_org": "Armin Haller, Krzysztof Janowicz, Simon Cox, Maxime Lefrancois, Kerry Taylor, Danh Le Phuoc, W3C Spatial Data on the Web Working Group", "year": 2017, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/vocab-ssn", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines SOSA/SSN ontologies for sensors, observations, samples, actuators, procedures, platforms, deployments, features of interest, and result semantics.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "iot", "knowledge-graph", "observations", "semantic-interoperability", "sosa", "ssn", "w3c" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "phase4-std-w3c-sparql-service-description-2013", "title": "SPARQL 1.1 Service Description", "authors_or_org": "Gregory Williams, W3C SPARQL Working Group", "year": 2013, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/sparql11-service-description", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines RDF vocabulary for describing SPARQL services, including supported entailment regimes, datasets, graph stores, features, functions, and endpoint capabilities.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "endpoint-governance", "metadata", "semantic-interoperability", "service-description", "sparql", "w3c" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oa-https-doi-org-10-5334-tilr-303", "title": "The New Public Analytics as an Emerging Paradigm in Public Sector Administration", "authors_or_org": "Karen Yeung", "year": 2022, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4322711967", "doi_or_identifier": "10.5334/tilr.303", "venue_or_site": "Tilburg Law Review", "abstract_or_summary": "The turn to data-driven approaches within public administration to inform (and even to automate) public sector decision-making can be understood as an emerging movement that I call the ‘New Public Analytics’ (‘NPA’). Central to the New Public Analytics is the use of data analytics a form of computational analysis that has its theoretical foundations in data science and statistics, involving the application of software algorithms (including but not limited to machine learning algorithms) to large data sets in order to identify patterns and correlations in the data capable of generating ‘actionable’ insight. The lecture will explore, amongst other things, the problematic and potentially dangerous pathologies of NPA, underpinning the need for lawyers to critically scrutinise these developments in order to identify ways in which law can be harnessed to ensure that adequate public accountability for NPA techniques is ensured.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 52 }, { "id": "oa-https-doi-org-10-1093-nar-gkaa1113", "title": "The Gene Ontology resource: enriching a GOld mine", "authors_or_org": "Seth Carbon, Eric Douglass, Benjamin M. Good, Deepak Unni, Nomi L. Harris, Chris Mungall, Siddartha Basu, Rex L. Chisholm", "year": 2020, "source_type": "paper", "bucket": "academic", "url": "https://openalex.org/W3112600195", "doi_or_identifier": "10.1093/nar/gkaa1113", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "The Gene Ontology Consortium (GOC) provides the most comprehensive resource currently available for computable knowledge regarding the functions of genes and gene products. Here, we report the advances of the consortium over the past two years. The new GO-CAM annotation framework was notably improved, and we formalized the model with a computational schema to check and validate the rapidly increasing repository of 2838 GO-CAMs. In addition, we describe the impacts of several collaborations to refine GO and report a 10% increase in the number of GO annotations, a 25% increase in annotated gene products, and over 9,400 new scientific articles annotated. As the project matures, we continue our efforts to review older annotations in light of newer findings, and, to maintain consistency with other ontologies. As a result, 20 000 annotations derived from experimental data were reviewed, corresponding to 2.5% of experimental GO annotations. The website (http://geneontology.org) was redesigned for quick access to documentation, downloads and tools. To maintain an accurate resource and support traceability and reproducibility, we have made available a historical archive covering the past 15 years of GO data with a consistent format and file structure for both the ontology and annotations.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "foundational", "openalex" ], "triage_tier": "candidate", "triage_score": 51 }, { "id": "s2-10-30574-wjaets-2026-19-1-0172", "title": "A multi-layer governance architecture for enterprise generative AI systems", "authors_or_org": "N. Ramachandran", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://www.semanticscholar.org/paper/2b6bf2d07e9ef4928003c29a0ed989fdf225b300", "doi_or_identifier": "10.30574/wjaets.2026.19.1.0172", "venue_or_site": "World Journal of Advanced Engineering Technology and Sciences", "abstract_or_summary": "Generative AI (GenAI)systems are increasingly embedded in enterprise workflows, yet existing governance approaches remain fragmented across policy, development, evaluation, and operational monitoring. GenAI systems produce unpredictable results because they generate output through probabilistic processes rather than deterministic logic, and large language models and foundation models have accelerated enterprise adoption across knowledge work, decision support, and content generation use cases. At the same time, generative systems introduce governance concerns related to hallucinations, output reliability, and accountability, which make enterprise deployment materially different from traditional software systems. Organizations commonly struggle to develop effective governance systems that oversee all stages of GenAI system development because these systems operate across multiple teams, technologies, data sources, and regulatory frameworks.\nThe study presents the EAGLE (Enterprise AI Governance and Lifecycle Execution) Framework, which functions as a multi-layer governance framework that assists organizations in establishing responsible and large-scale Generative AI deployment. The framework includes four key layers: Governance, which sets policies for responsible AI use, compliance, and data management; Program Orchestration, which coordinates collaboration between engineering, infrastructure, security, legal, and product teams; Evaluation, which validates model performance, output reliability, and business impact; and Operational Monitoring, which continuously tracks system performance, detects model drift, and manages incidents. The research team used Design Science Research to create and test the EAGLE framework which functions as an organized governance system for enterprise AI implementation. The research results demonstrate that EAGLE multi-layer governance systems enable organizations to prepare for extensive AI implementation through improved risk management capabilities and better accountability systems and enhanced interdepartmental cooperation. The system provides organizations with an operational guide that enables them to implement Generative AI systems in a dependable and regulation-compliant manner.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "commercial", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 50 }, { "id": "s2-10-1109-access-2026-3702066", "title": "A Review of Neuro-Symbolic AI for Smart Cities: Architectures, Ontologies, and IoT-Driven Applications", "authors_or_org": "A. Pliatsios, Panagiotis Mpatos, Michael Dossis", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://www.semanticscholar.org/paper/26ca9e17166ddf809fa2142520d4734514b14872", "doi_or_identifier": "10.1109/access.2026.3702066", "venue_or_site": "IEEE Access", "abstract_or_summary": "Smart cities rely on Internet of Things (IoT) ecosystems and Artificial Intelligence (AI) to manage complex urban infrastructures. While deep learning provides predictive power, it often lacks the transparency and semantic awareness required for safety-critical governance. This paper reviews the state-of-the-art in Neuro-Symbolic AI (NeSy AI) for smart cities, examining its capacity to bridge data-driven perception with structured, logic-based reasoning. Following a systematic literature review (2020–2026), we classify neuro-symbolic architectures into sequential pipelines, differentiable reasoning models, and knowledge graph-driven frameworks. We analyze applications across transportation, energy, and public safety, highlighting the critical role of ontology integration and explainable AI mechanisms. To address identified scalability and semantic drift bottlenecks, we propose a novel, unified Edge-to-Cloud conceptual framework. This review serves as a roadmap for researchers and practitioners, demonstrating that neuro-symbolic architectures provide a robust foundation for achieving auditable, ethically aligned, and semantically grounded urban intelligence.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "ontology_ai", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 50 }, { "id": "oa-https-doi-org-10-1111-anti-12641", "title": "Platform Capitalism’s Hidden Abode: Producing Data Assets in the Gig Economy", "authors_or_org": "Niels van Doorn, Adam Badger", "year": 2020, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W3033766131", "doi_or_identifier": "10.1111/anti.12641", "venue_or_site": "Antipode", "abstract_or_summary": "Abstract In this article, we argue that the governance of gig work under conditions of financialised platform capitalism is characterised by a process that we call “dual value production”: the monetary value produced by the service provided is augmented by the use and speculative value of the data produced before, during, and after service provision. App‐governed gig workers hence function as pivotal conduits in software systems that produce digital data as a particular asset class. We reflect on the production of data assets and the unequal distribution of opportunities for their valorisation, after which we survey a number of strategies seeking data‐centric worker empowerment. These strategies, we argue, are crucial attempts to push back against platform capitalism’s domination, bankrolled by what we term “meta‐platforms”. Ultimately, it is the massive wealth and synergetic capacities of meta‐platforms that constitute the most formidable obstacle to worker power and social justice in increasingly data‐driven societies.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 49 }, { "id": "arxiv-2511-06455v1", "title": "A Multi-Agent System for Semantic Mapping of Relational Data to Knowledge Graphs", "authors_or_org": "Milena Trajanoska, Riste Stojanov, Dimitar Trajanov", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2511.06455", "doi_or_identifier": "arxiv:2511.06455", "venue_or_site": "arXiv", "abstract_or_summary": "Enterprises often maintain multiple databases for storing critical business data in siloed systems, resulting in inefficiencies and challenges with data interoperability. A key to overcoming these challenges lies in integrating disparate data sources, enabling businesses to unlock the full potential of their data. Our work presents a novel approach for integrating multiple databases using knowledge graphs, focusing on the application of large language models as semantic agents for mapping and connecting structured data across systems by leveraging existing vocabularies. The proposed methodology introduces a semantic layer above tables in relational databases, utilizing a system comprising multiple LLM agents that map tables and columns to Schema.org terms. Our approach achieves a mapping accuracy of over 90% in multiple domains.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "enterprise-data", "knowledge-graph-construction", "llm-agents", "ontology_ai", "relational-data", "schema-org", "semantic-mapping" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-floridi-cowls-2019-unified-framework", "title": "A Unified Framework of Five Principles for AI in Society", "authors_or_org": "Luciano Floridi and Josh Cowls", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1162/99608f92.8cd550d1", "doi_or_identifier": "10.1162/99608f92.8cd550d1", "venue_or_site": "Harvard Data Science Review", "abstract_or_summary": "AI ethics framework synthesizing beneficence, non-maleficence, autonomy, justice, and explicability.", "key_claims": [ "AI governance can be organized around a small set of ethical principles.", "Explicability combines intelligibility and accountability.", "Principles require translation into concrete practices, standards, and oversight." ], "ontology_relevance": "Useful for mapping ontology governance requirements to ethical principles such as explicability, justice, and autonomy.", "ai_relevance": "Widely cited AI ethics synthesis for article background, especially when paired with critiques of principles-only governance.", "palantir_relevance": "Helpful neutral vocabulary for discussing operational AI governance in enterprise and public-sector settings.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-ethics", "explicability", "governance", "phase3", "principles", "society" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "arxiv-2511-11017v1", "title": "AI Agent-Driven Framework for Automated Product Knowledge Graph Construction in E-Commerce", "authors_or_org": "Dimitar Peshevski, Riste Stojanov, Dimitar Trajanov", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2511.11017v1", "doi_or_identifier": "2511.11017v1", "venue_or_site": "arXiv", "abstract_or_summary": "The rapid expansion of e-commerce platforms generates vast amounts of unstructured product data, creating significant challenges for information retrieval, recommendation systems, and data analytics. Knowledge Graphs (KGs) offer a structured, interpretable format to organize such data, yet constructing product-specific KGs remains a complex and manual process. This paper introduces a fully automated, AI agent-driven framework for constructing product knowledge graphs directly from unstructured product descriptions. Leveraging Large Language Models (LLMs), our method operates in three stages using dedicated agents: ontology creation and expansion, ontology refinement, and knowledge graph population. This agent-based approach ensures semantic coherence, scalability, and high-quality output without relying on predefined schemas or handcrafted extraction rules. We evaluate the system on a real-world dataset of air conditioner product descriptions, demonstrating strong performance in both ontology generation and KG population. The framework achieves over 97\\% property coverage and minimal redundancy, validating its effectiveness and practical applicability. Our work highlights the potential of LLMs to automate structured knowledge extraction in retail, providing a scalable path toward intelligent product data integration and utilization.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "s2-2605-29168", "title": "Better Later Than Sooner: Neuro-Symbolic Knowledge Graph Construction via Ontology-grounded Post-extraction Correction", "authors_or_org": "Lorenzo Loconte, T. Hospedales, Cristina Cornelio", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2605.29168v1", "doi_or_identifier": "2605.29168v1", "venue_or_site": "Semantic Scholar", "abstract_or_summary": "Question answering (QA) is a core challenge in AI, particularly for complex queries requiring multi-hop reasoning across documents, or symbolic operations like aggregation or exhaustive listing. Retrieval-augmented generation has become the dominant approach to QA, with recent graph-based variants addressing part of these issues by organizing knowledge to better support compositional questions. However, most textual graph-based RAG methods still lack the structure needed for symbolic operations useful to answer complex questions reliably. This motivates symbolic graph-based approaches, which extract knowledge graphs (KGs) whose relations are logic predicates that enable SQL-like querying. Yet these pipelines typically use LLMs for KG extraction, which can introduce consistency issues, where extracted facts may violate commonsense ontology constraints. We propose a neuro-symbolic framework for ontology-grounded KG construction combining open-domain extraction, embedding-based canonicalization of types and predicates, and targeted LLM-based correction of ontology violations. By deferring corrections to a post-extraction stage, our method avoids repeated LLM calls, substantially reducing token usage while improving KG consistency and preserving downstream QA quality. Finally, we show that the extracted KGs are well suited for symbolic querying by measuring the occurrence of SPARQL graph patterns.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-borgman-2015-big-data-little-data", "title": "Big Data, Little Data, No Data: Scholarship in the Networked World", "authors_or_org": "Christine L. Borgman", "year": 2015, "source_type": "book", "bucket": "books", "url": "https://mitpress.mit.edu/9780262529914/big-data-little-data-no-data", "doi_or_identifier": "isbn 9780262529914", "venue_or_site": "MIT Press", "abstract_or_summary": "Scholarly infrastructure book on data sharing, curation, metadata, stewardship, and research data practices.", "key_claims": [ "Data reuse depends on context, metadata, documentation, and stewardship.", "Infrastructure for data is social as much as technical.", "Data sharing and interoperability require incentives, governance, and ongoing maintenance." ], "ontology_relevance": "Supports the idea that ontology value depends on stewardship, metadata, and reuse practices rather than model elegance alone.", "ai_relevance": "AI-ready knowledge bases require curation, provenance, documentation, and incentives for maintenance.", "palantir_relevance": "Useful for evaluating enterprise ontology as durable data stewardship infrastructure.", "quality_signal": "widely_cited", "retrieval_tags": [ "data-sharing", "data-stewardship", "maintenance", "metadata", "phase3", "research-infrastructure" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "noy_etal_2009_bioportal", "title": "BioPortal: Ontologies and Integrated Data Resources at the Click of a Mouse", "authors_or_org": "Natalya F. Noy, Nigam H. Shah, Patricia L. Whetzel, et al.", "year": 2009, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/nar/gkp440", "doi_or_identifier": "10.1093/nar/gkp440", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "Describes BioPortal, a web-based repository for biomedical ontologies, metadata, mappings, search, visualization, and services.", "key_claims": [ "Ontology repositories need metadata, access services, search, visualization, and mappings.", "Community ontology reuse improves when ontologies are discoverable and service-accessible.", "Ontology ecosystems require infrastructure beyond individual ontology files." ], "ontology_relevance": "Important example of ontology repository and service infrastructure.", "ai_relevance": "Shows how ontology services can support retrieval, annotation, entity normalization, and AI data pipelines.", "palantir_relevance": "Relevant to enterprise ontology catalogs, discovery, mapping, and governance services.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical", "bioportal", "mappings", "ontology-repository", "services" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "arxiv-2603-08321v1", "title": "CORE-Acu: Structured Reasoning Traces and Knowledge Graph Safety Verification for Acupuncture Clinical Decision Support", "authors_or_org": "Liuyi Xu, Yun Guo, Ming Chen, Zihan Dun, Yining Qian, An-Yang Lu, Shuang Li, Lijun Liu", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2603.08321v1", "doi_or_identifier": "2603.08321v1", "venue_or_site": "arXiv", "abstract_or_summary": "Large language models (LLMs) show significant potential for clinical decision support (CDS), yet their black-box nature -- characterized by untraceable reasoning and probabilistic hallucinations -- poses severe challenges in acupuncture, a field demanding rigorous interpretability and safety. To address this, we propose CORE-Acu, a neuro-symbolic framework for acupuncture clinical decision support that integrates Structured Chain-of-Thought (S-CoT) with knowledge graph (KG) safety verification. First, we construct the first acupuncture Structured Reasoning Trace dataset and a schema-constrained fine-tuning framework. By enforcing an explicit causal chain from pattern identification to treatment principles, treatment plans, and acupoint selection, we transform implicit Traditional Chinese Medicine (TCM) reasoning into interpretable generation constraints, mitigating the opacity of LLM-based CDS. Furthermore, we construct a TCM safety knowledge graph and establish a ``Generate--Verify--Revise'' closed-loop inference system based on a Symbolic Veto Mechanism, employing deterministic rules to intercept hallucinations and enforce hard safety boundaries. Finally, we introduce the Lexicon-Matched Entity-Reweighted Loss (LMERL), which corrects terminology drift caused by the frequency--importance mismatch in general optimization by adaptively amplifying gradient contributions of high-risk entities during fine-tuning. Experiments on 1,000 held-out cases demonstrate CORE-Acu's superior entity fidelity and reasoning quality. Crucially, CORE-Acu achieved 0/1,000 observed safety violations (95\\% CI: 0--0.37\\%), whereas GPT-4o exhibited an 8.5\\% violation rate under identical rules. These results establish CORE-Acu as a robust neuro-symbolic framework for acupuncture clinical decision support, guaranteeing both reasoning auditability and strict safety compliance.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "p2-llm-ont-036", "title": "Data Products Ontology Specification Version 1.0 beta", "authors_or_org": "Object Management Group", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://www.omg.org/spec/DPROD/1.0/Beta1/About-DPROD", "doi_or_identifier": "omg:ptc/25-02-01", "venue_or_site": "Object Management Group", "abstract_or_summary": "OMG beta specification for DPROD, a DCAT profile and ontology for describing data products, ports, data services, metadata, and marketplace-oriented discovery.", "key_claims": [ "Data products need shared machine-readable semantics beyond table schemas.", "DPROD extends DCAT to describe data products and services in decentralized marketplaces.", "SHACL constraints can support conformance for data product metadata." ], "ontology_relevance": "High-value emerging standard for semantic data products and enterprise data marketplace interoperability.", "ai_relevance": "Gives LLM agents structured metadata for discovering, selecting, and reasoning over data products.", "palantir_relevance": "Direct comparator to enterprise ontology layers that treat data products as governed operational assets.", "quality_signal": "official_docs", "retrieval_tags": [ "data-marketplace", "data-product", "dcat", "dprod", "omg", "ontology", "semantic-contracts", "semantic-layer" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase7-deepgo-ontology-aware-classifier-2018", "title": "DeepGO: predicting protein functions from sequence and interactions using a deep ontology-aware classifier", "authors_or_org": "Maxat Kulmanov; Mohammed Asif Khan; Robert Hoehndorf", "year": 2018, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/bioinformatics/btx624", "doi_or_identifier": "10.1093/bioinformatics/btx624", "venue_or_site": "Bioinformatics", "abstract_or_summary": "DeepGO predicts protein functions from sequence and protein-protein interactions using a deep ontology-aware classifier based on Gene Ontology structure.", "key_claims": [ "Gene Ontology structure can guide deep-learning protein function prediction.", "Ontology-aware classifiers can exploit hierarchical biological function labels.", "AI prediction tasks can use ontology as supervision and output space." ], "ontology_relevance": "Direct evidence for ontology as an AI modeling substrate, not just a post-hoc annotation vocabulary.", "ai_relevance": "Early example of deep learning directly using ontology structure for supervised biological prediction.", "palantir_relevance": "Conceptually relevant to ontology-constrained model outputs in enterprise domains.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ai-for-science", "deep-learning", "gene-ontology", "ontology-aware-ai", "phase7", "protein-function" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase7-deepgoplus-protein-function-2019", "title": "DeepGOPlus: improved protein function prediction from sequence", "authors_or_org": "Maxat Kulmanov; Robert Hoehndorf", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/bioinformatics/btz595", "doi_or_identifier": "10.1093/bioinformatics/btz595", "venue_or_site": "Bioinformatics", "abstract_or_summary": "DeepGOPlus improves protein function prediction from sequence using deep learning and Gene Ontology annotations, combining sequence-based predictions with similarity-based approaches.", "key_claims": [ "Protein function prediction can use Gene Ontology as the structured output vocabulary.", "Combining deep learning with similarity-based methods improves function prediction in the reported evaluation.", "Ontology-based annotation resources are central to AI-for-biology benchmarks." ], "ontology_relevance": "Shows how scientific ontologies become machine-learning supervision and evaluation infrastructure.", "ai_relevance": "Demonstrates ontology-backed label spaces and benchmarks for AI in protein function prediction.", "palantir_relevance": "Analogous to enterprise ontology labels constraining prediction tasks and downstream decision support.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ai-for-science", "deep-learning", "gene-ontology", "ontology-aware-ai", "phase7", "protein-function" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase11-joshi-2026-deontic-policies-agentic-ai-runtime-governance", "title": "Deontic Policies for Runtime Governance of Agentic AI Systems", "authors_or_org": "Anupam Joshi; Tim Finin; Karuna Pande Joshi; Lalana Kagal", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.19464", "doi_or_identifier": "10.48550/arxiv.2606.19464", "venue_or_site": "arXiv", "abstract_or_summary": "Introduces AgenticRei, a deontic policy approach for runtime governance of LLM-driven agentic systems. The framework uses an OWL-expressed Rei policy language evaluated outside the LLM to handle permissions, prohibitions, obligations, dispensations, policy precedence, ontological class reasoning, tool invocations, and agent-to-agent messages.", "key_claims": [ "Tool-using LLM agents create governance needs beyond authentication and access control.", "AgenticRei models permissions, prohibitions, obligations, dispensations, conflict resolution, and ontological reasoning over domain classes.", "The policy engine is evaluated outside the LLM and governs both tool calls and agent-to-agent messages.", "Existing engines such as XACML, Rego, and Cedar are framed as covering only part of the governance structure needed for autonomous agents." ], "ontology_relevance": "Strong evidence that ontology and OWL-style policy reasoning can become runtime governance infrastructure for agentic AI.", "ai_relevance": "High-value frontier source for agentic AI governance, runtime policy enforcement, and tool-use control outside the LLM.", "palantir_relevance": "Provides an independent research comparator for Palantir Ontology MCP, dynamic security, action permissions, and external-agent tool governance.", "quality_signal": "preprint", "retrieval_tags": [ "agentic-ai", "deontic-policy", "governed-action", "llm", "ontology-governance", "owl", "phase11", "policy-reasoning", "runtime-governance", "tool-use" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-llm-ontmem-004", "title": "Dynamic Retrieval Augmented Generation of Ontologies using Artificial Intelligence (DRAGON-AI)", "authors_or_org": "Sabrina Toro; Anna V. Anagnostopoulos; Sue Bello; Kai Blumberg; Rhiannon Cameron; Leigh Carmody; Alexander D. Diehl; Damion Dooley; William Duncan; Petra Fey; Pascale Gaudet; Nomi L. Harris; Marcin Joachimiak; Leila Kiani; Tiago Lubiana; Monica C. Munoz-Torres; Shawn O'Neil; David Osumi-Sutherland; Aleix Puig; Justin P. Reese; Leonore Reiser; Sofia Robb; Troy Ruemping; James Seager; Eric Sid; Ray Stefancsik; Magalie Weber; Valerie Wood; Melissa A. Haendel; Christopher J. Mungall", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://jbiomedsem.biomedcentral.com/articles/10.1186/s13326-024-00325-z", "doi_or_identifier": "10.1186/s13326-024-00325-z", "venue_or_site": "Journal of Biomedical Semantics", "abstract_or_summary": "Describes DRAGON-AI, a RAG-based method for generating textual and logical ontology components from existing ontologies and text sources, evaluated on de novo term construction across biomedical and environmental ontologies.", "key_claims": [ "RAG can help generate ontology terms by retrieving relevant existing ontology and text context.", "Expert evaluators found useful relationship generation, but human-authored definitions remained stronger in some dimensions.", "Domain experts with higher confidence were better at detecting flaws in AI-generated ontology content." ], "ontology_relevance": "High-value biomedical evidence for RAG-assisted ontology construction with expert review.", "ai_relevance": "Shows both promise and limits of LLM/RAG systems for formal and textual ontology components.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "dragon-ai", "expert-evaluation", "ontology-generation", "rag" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "p2-llm-ont-011", "title": "Enriching Ontologies with Disjointness Axioms using Large Language Models", "authors_or_org": "Elias Crum; Antonio De Santis; Manon Ovide; Jiaxin Pan; Alessia Pisu; Nicolas Lazzari; Sebastian Rudolph", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2410.03235", "doi_or_identifier": "arxiv:2410.03235", "venue_or_site": "KBC-LM 2024 at ISWC 2024 / CEUR-WS / arXiv", "abstract_or_summary": "Uses LLMs to propose class disjointness axioms, with logical checks to preserve satisfiability and reduce unnecessary LLM calls.", "key_claims": [ "LLMs can propose useful disjointness axioms when guided by prompt strategies.", "Logical dependencies between subclass and disjointness statements must be handled explicitly.", "Automated ontology enrichment should include satisfiability-preserving checks." ], "ontology_relevance": "Moves beyond taxonomy extraction into logical axiom enrichment and consistency-aware ontology completion.", "ai_relevance": "Combines LLM suggestions with symbolic validation to control output quality.", "palantir_relevance": "Relevant to adding negative constraints and consistency checks to enterprise object models.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "disjointness", "iswc-workshop", "llm", "logical-consistency", "ontology-enrichment" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase6-zhou-2025-kg-rag-incompleteness", "title": "Evaluating Knowledge Graph Based Retrieval Augmented Generation Methods under Knowledge Incompleteness", "authors_or_org": "Dongzhuoran Zhou; Yuqicheng Zhu; Yuan He; Jiaoyan Chen; Evgeny Kharlamov; Steffen Staab", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2504.05163", "doi_or_identifier": "10.48550/arxiv.2504.05163", "venue_or_site": "arXiv", "abstract_or_summary": "Systematically evaluates how incomplete knowledge graphs affect KG-RAG methods by removing triples and measuring downstream question-answering effects.", "key_claims": [ "KG-RAG systems are sensitive to knowledge graph incompleteness.", "Existing benchmarks underrepresent missing-knowledge conditions.", "Robustness to incomplete KGs should be part of KG-RAG evaluation." ], "ontology_relevance": "Relevant to ontology/KG coverage, completeness, and schema-aware evaluation risks.", "ai_relevance": "Evaluates retrieval-grounded LLM QA under realistic KG defects.", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "evaluation", "kg-rag", "knowledge-graph", "knowledge-incompleteness", "phase6", "qa-robustness", "rag" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "guarino_welty_2002_ontoclean", "title": "Evaluating Ontological Decisions with OntoClean", "authors_or_org": "Nicola Guarino and Christopher A. Welty", "year": 2002, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/503124.503150", "doi_or_identifier": "10.1145/503124.503150", "venue_or_site": "Communications of the ACM", "abstract_or_summary": "Presents OntoClean, a methodology for evaluating taxonomic choices using meta-properties such as rigidity, identity, unity, and dependence.", "key_claims": [ "Taxonomic hierarchies can contain subtle ontological errors.", "Meta-properties can expose mistaken subclass relations.", "Identity, unity, dependence, and rigidity matter for high-quality category systems.", "Ontology evaluation must include conceptual discipline, not only syntax or graph completeness.", "OntoClean uses meta-properties such as rigidity, identity, unity, and dependence to detect taxonomic modeling errors.", "AI systems that inherit bad taxonomies can make category mistakes in retrieval, classification, and workflow automation." ], "ontology_relevance": "Important quality-control method for ontology taxonomies.", "ai_relevance": "Prevents category mistakes that can propagate through AI classification, retrieval, and reasoning systems.", "palantir_relevance": "Relevant to enterprise object models where roles, states, physical objects, and events are often confused.", "quality_signal": "widely_cited", "retrieval_tags": [ "conceptual-quality", "identity", "ontoclean", "ontology-evaluation", "ontology-governance", "phase16", "rigidity", "taxonomy" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase2-he-2023-llm-ontology-alignment", "title": "Exploring Large Language Models for Ontology Alignment", "authors_or_org": "Yuan He; Jiaoyan Chen; Hang Dong; Ian Horrocks", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2309.07172", "doi_or_identifier": "arxiv:2309.07172", "venue_or_site": "arXiv", "abstract_or_summary": "Investigates zero-shot use of generative LLMs for ontology alignment on challenging OAEI Bio-ML equivalence matching subsets with labels and structural context.", "key_claims": [ "LLMs can be competitive in ontology alignment when concept labels and structural contexts are carefully represented.", "Prompt and framework design strongly affect LLM alignment performance.", "Ontology alignment is a concrete testbed for measuring hallucination and semantic precision in LLMs." ], "ontology_relevance": "Important early LLM ontology alignment paper from KR/Semantic Web researchers.", "ai_relevance": "Tests LLM semantic matching capabilities on formal ontology tasks.", "palantir_relevance": "Relevant to enterprise schema alignment and ontology mapping problems.", "quality_signal": "preprint", "retrieval_tags": [ "evaluation", "llm", "oaei", "ontology-alignment", "ontology-matching", "schema-alignment" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase2-li-2024-graphreader", "title": "GraphReader: Building Graph-based Agent to Enhance Long-Context Abilities of Large Language Models", "authors_or_org": "Shilong Li; Yancheng He; Hangyu Guo; Xingyuan Bu; Ge Bai; Jie Liu; Jiaheng Liu; Xingwei Qu; Yangguang Li; Wanli Ouyang; Wenbo Su; Bo Zheng", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2406.14550", "doi_or_identifier": "arxiv:2406.14550", "venue_or_site": "arXiv", "abstract_or_summary": "Presents a graph-based agent system that structures long texts into a graph and lets an agent explore nodes and neighbors through predefined functions.", "key_claims": [ "Graph navigation can substitute for very long context windows in some long-document tasks.", "Predefined read-neighbor functions give agents controlled access to structured evidence.", "Graph-based exploration supports coarse-to-fine retrieval and iterative reflection.", "Graph navigation can reduce the burden of placing all long-context evidence directly into the prompt.", "Agentic traversal helps locate relevant evidence across long documents.", "Graph-based reading needs reliable node construction and stopping criteria.", "Graph navigation can extend effective long-context reasoning without placing all text in the prompt.", "Agentic tools for reading nodes and neighbors make retrieval an interactive process.", "Graph-based exploration supports both single-hop and multi-hop question answering." ], "ontology_relevance": "Connects ontology-like graph navigation to agentic retrieval and long-context reasoning.", "ai_relevance": "Useful for agent architectures that traverse structured memory rather than ingesting all text.", "palantir_relevance": "Comparable to agents traversing Palantir object/link structures via constrained tools.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "agent-memory", "document-graph", "emnlp-2024", "graph-agent", "graph-navigation", "graph-reader", "graph-retrieval", "graphreader", "llm-agent", "llm-agents", "long-context", "retrieval", "tool-use" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase2-graphrag-019", "title": "iText2KG: Incremental Knowledge Graphs Construction Using Large Language Models", "authors_or_org": "Yassir Lairgi; Ludovic Moncla; Rémy Cazabet; Khalid Benabdeslem; Pierre Cléau", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2409.03284", "doi_or_identifier": "arxiv:2409.03284", "venue_or_site": "arXiv", "abstract_or_summary": "Proposes an incremental LLM-based pipeline for constructing knowledge graphs from text, focusing on repeated extraction, updating, and graph growth.", "key_claims": [ "Incremental KG construction is necessary when corpora evolve over time.", "LLM extraction pipelines need deduplication, alignment, and update policies.", "Graph construction quality determines whether later GraphRAG retrieval is trustworthy." ], "ontology_relevance": "Relevant to maintaining ontology-backed graphs as living knowledge bases rather than one-time extraction outputs.", "ai_relevance": "Useful for the graph construction side of GraphRAG and KAG systems.", "palantir_relevance": "Relevant to operational knowledge bases that must track changing events, entities, and documents.", "quality_signal": "preprint", "retrieval_tags": [ "graph-construction", "incremental-kg", "itext2kg", "llm-extraction", "updates" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "studer_benjamins_fensel_1998_knowledge_engineering", "title": "Knowledge Engineering: Principles and Methods", "authors_or_org": "Rudi Studer, V. Richard Benjamins, and Dieter Fensel", "year": 1998, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/S0169-023X%2897%2900056-6", "doi_or_identifier": "10.1016/s0169-023x(97)00056-6", "venue_or_site": "Data & Knowledge Engineering", "abstract_or_summary": "Synthesizes knowledge engineering methods and gives the influential formulation of ontologies as formal, explicit specifications of shared conceptualizations.", "key_claims": [ "Ontology engineering sits inside the broader lifecycle of knowledge engineering.", "An ontology should be formal, explicit, and shared when used for system interoperability.", "Knowledge acquisition, modeling, reuse, and maintenance are connected tasks." ], "ontology_relevance": "Canonical source for the formal-explicit-shared-conceptualization definition.", "ai_relevance": "Connects ontology work to reusable knowledge assets for AI systems.", "palantir_relevance": "Relevant to enterprise ontology as a shared model among data, applications, and human workflows.", "quality_signal": "widely_cited", "retrieval_tags": [ "knowledge-engineering", "ontology-definition", "reuse", "shared-conceptualization" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-llm-ontmem-026", "title": "Knowledge Graph-based Retrieval-Augmented Generation for Schema Matching", "authors_or_org": "Chuangtao Ma; Sriom Chakrabarti; Arijit Khan; Balint Molnar", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2501.08686", "doi_or_identifier": "arxiv:2501.08686", "venue_or_site": "arXiv", "abstract_or_summary": "Investigates knowledge graph-based RAG for schema matching, using graph context to improve retrieval and matching of schema elements.", "key_claims": [ "Knowledge graph context can improve schema matching by retrieving related semantic evidence around candidate elements.", "Schema matching for data integration benefits from combining LLM reasoning with graph-based retrieval.", "KG-RAG for matching highlights the convergence of data integration, retrieval, and semantic alignment." ], "ontology_relevance": "Directly links KG retrieval techniques to schema and ontology alignment tasks.", "ai_relevance": "Shows RAG as an alignment aid, not only an answering mechanism.", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "data-integration", "graph-retrieval", "kg-rag", "ontology-alignment", "schema-matching" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "p2-llm-ont-006", "title": "Large Language Models for Scholarly Ontology Generation: An Extensive Analysis in the Engineering Field", "authors_or_org": "Tanay Aggarwal; Angelo Salatino; Francesco Osborne; Enrico Motta", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.ipm.2025.104262", "doi_or_identifier": "10.1016/j.ipm.2025.104262", "venue_or_site": "Information Processing & Management / arXiv", "abstract_or_summary": "Evaluates multiple LLMs and prompting strategies for identifying semantic relations among engineering research topics against a gold standard based on the IEEE Thesaurus.", "key_claims": [ "LLMs can classify broader, narrower, same-as, and other topic relations with strong performance under some prompting strategies.", "Smaller or quantized models may be competitive when prompts are tuned for the task.", "Scholarly ontologies need relation-specific evaluation rather than generic text similarity scores.", "LLMs can support relationship classification among scholarly concepts.", "Scholarly ontology generation remains sensitive to relation semantics and evaluation design.", "Engineering-domain analysis provides a concrete domain testbed." ], "ontology_relevance": "Shows LLM-based ontology generation in a scholarly domain with a controlled reference vocabulary.", "ai_relevance": "Useful for comparing model families and prompting methods on semantic relation classification.", "palantir_relevance": "Relevant to building topic and capability taxonomies over technical corpora or engineering assets.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "engineering", "engineering-domain", "information-processing-management", "ipm-2026", "llm", "ontology-generation", "phase6", "relation-classification", "scholarly-ontology", "semantic-relations" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase11-oyewale-soru-2026-ontoekg-enterprise-ontology-construction", "title": "LLM-Driven Ontology Construction for Enterprise Knowledge Graphs", "authors_or_org": "Abdulsobur Oyewale; Tommaso Soru", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2602.01276", "doi_or_identifier": "10.48550/arxiv.2602.01276", "venue_or_site": "arXiv / ICSC 2026 preprint", "abstract_or_summary": "Introduces OntoEKG, an LLM-driven pipeline for constructing enterprise knowledge-graph ontologies from unstructured enterprise documents. The approach separates class/property extraction from entailment-based hierarchy construction and reports both promise and limitations, including domain F1 variation and hierarchy/scope challenges.", "key_claims": [ "Enterprise knowledge graphs require ontologies for semantic governance over heterogeneous data.", "OntoEKG decomposes LLM-driven ontology construction into extraction and entailment modules before serializing standard RDF.", "Reported results show promise while exposing limitations in scope definition and hierarchical reasoning.", "The paper highlights a shortage of comprehensive benchmarks for end-to-end enterprise ontology construction." ], "ontology_relevance": "Directly addresses ontology generation for enterprise knowledge graphs, one of the project's most important current-literature gaps.", "ai_relevance": "Current preprint evidence for LLM-assisted enterprise ontology construction and end-to-end ontology benchmarking.", "palantir_relevance": "Useful non-Palantir comparator for enterprise ontology automation; it supports a broader enterprise ontology trend without validating Palantir claims.", "quality_signal": "preprint", "retrieval_tags": [ "enterprise-knowledge-graph", "enterprise-ontology", "llm", "ontology-construction", "ontology-engineering", "ontology-generation", "phase11", "rdf", "semantic-governance" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase2-llms4life-2024", "title": "LLMs4Life: Large Language Models for Ontology Learning in Life Sciences", "authors_or_org": "Nadeen Fathallah; Steffen Staab; Alsayed Algergawy", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2412.02035", "doi_or_identifier": "arxiv:2412.02035", "venue_or_site": "arXiv / ELMKE", "abstract_or_summary": "Extends LLM-based ontology learning to life-science domains using prompt engineering, ontology reuse, and evaluation of logical consistency, completeness, and scalability.", "key_claims": [ "Specialized scientific domains expose limits in LLM ontology learning around hierarchy depth and class coverage.", "Ontology reuse and prompt pipelines can improve domain-specific ontology generation.", "Evaluation needs to cover logical consistency, completeness, scalability, and structural quality." ], "ontology_relevance": "Shows LLM ontology learning challenges in a high-stakes scientific domain.", "ai_relevance": "Useful for Nature-style framing around scientific ontologies and LLM reliability.", "palantir_relevance": "Relevant to healthcare/public-sector ontology use cases, but not Palantir-specific.", "quality_signal": "preprint", "retrieval_tags": [ "biomedical-ontology", "life-sciences", "llm", "ontology-learning", "reuse", "validation" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "arxiv-2109-07401v1", "title": "Matching with Transformers in MELT", "authors_or_org": "Sven Hertling, Jan Portisch, Heiko Paulheim", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2109.07401v1", "doi_or_identifier": "2109.07401v1", "venue_or_site": "arXiv", "abstract_or_summary": "One of the strongest signals for automated matching of ontologies and knowledge graphs are the textual descriptions of the concepts. The methods that are typically applied (such as character- or token-based comparisons) are relatively simple, and therefore do not capture the actual meaning of the texts. With the rise of transformer-based language models, text comparison based on meaning (rather than lexical features) is possible. In this paper, we model the ontology matching task as classification problem and present approaches based on transformer models. We further provide an easy to use implementation in the MELT framework which is suited for ontology and knowledge graph matching. We show that a transformer-based filter helps to choose the correct correspondences given a high-recall alignment and already achieves a good result with simple alignment post-processing methods.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "foundational" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase11-wu-2026-memgraphrag-memory-multi-agent-graphrag", "title": "MemGraphRAG: Memory-based Multi-Agent System for Graph Retrieval-Augmented Generation", "authors_or_org": "Chuanjie Wu; Zhishang Xiang; Yunbo Tang; Zerui Chen; Qinggang Zhang; Jinsong Su", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.00610", "doi_or_identifier": "10.48550/arxiv.2606.00610", "venue_or_site": "arXiv / KDD 2026 accepted preprint", "abstract_or_summary": "Proposes MemGraphRAG, a memory-based multi-agent framework for GraphRAG that uses shared memory to maintain global context during graph construction and a memory-aware hierarchical retrieval algorithm. The authors argue this reduces thematically inconsistent, logically conflicting, and structurally fragmented graphs in large unstructured corpora.", "key_claims": [ "GraphRAG can capture structural relationships for complex reasoning, but fragment-level graph extraction can create inconsistent or disconnected graphs.", "Shared memory across agents gives graph construction a global corpus context.", "Memory-aware hierarchical retrieval is proposed to exploit the constructed graph for retrieval-augmented generation.", "The paper reports improvements over baseline models while keeping efficiency comparable." ], "ontology_relevance": "Supports the claim that graph construction quality, consistency, and retrieval structure are central bottlenecks for ontology/KG-backed AI.", "ai_relevance": "Frontier evidence for multi-agent graph construction, graph memory, and RAG architectures that try to improve retrieval quality beyond fragment-level extraction.", "palantir_relevance": "Useful comparator for agent memory and graph retrieval; it is not Palantir-specific evidence.", "quality_signal": "preprint", "retrieval_tags": [ "agent-memory", "graph-based-memory", "graphrag", "knowledge-graph", "llm", "multi-agent", "phase11", "rag" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "fernandez_lopez_gomez_perez_juristo_1997_methontology", "title": "METHONTOLOGY: From Ontological Art Towards Ontological Engineering", "authors_or_org": "Mariano Fernandez-Lopez, Asuncion Gomez-Perez, and Natalia Juristo", "year": 1997, "source_type": "paper", "bucket": "academic", "url": "https://aaai.org/papers/0005-ss97-06-005-methontology-from-ontological-art-towards-ontological-engineering", "doi_or_identifier": "aaai ss-97-06-005", "venue_or_site": "AAAI Spring Symposium Series", "abstract_or_summary": "Presents METHONTOLOGY, an ontology engineering methodology with lifecycle activities for specification, conceptualization, formalization, implementation, evaluation, maintenance, and documentation.", "key_claims": [ "Ontology development should follow an explicit lifecycle.", "Evaluation and documentation are core engineering activities.", "Methodological discipline helps move ontology construction from craft to engineering." ], "ontology_relevance": "Canonical ontology engineering methodology source.", "ai_relevance": "AI-era ontologies require maintainable lifecycle practices as models and data sources change.", "palantir_relevance": "Relevant to governing operational ontologies as evolving enterprise assets rather than static schemas.", "quality_signal": "widely_cited", "retrieval_tags": [ "lifecycle", "methodology", "methontology", "ontology-engineering" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "arxiv-2308-04814v1", "title": "Neuro-Symbolic RDF and Description Logic Reasoners: The State-Of-The-Art and Challenges", "authors_or_org": "Gunjan Singh, Sumit Bhatia, Raghava Mutharaju", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2308.04814v1", "doi_or_identifier": "2308.04814v1", "venue_or_site": "arXiv", "abstract_or_summary": "Ontologies are used in various domains, with RDF and OWL being prominent standards for ontology development. RDF is favored for its simplicity and flexibility, while OWL enables detailed domain knowledge representation. However, as ontologies grow larger and more expressive, reasoning complexity increases, and traditional reasoners struggle to perform efficiently. Despite optimization efforts, scalability remains an issue. Additionally, advancements in automated knowledge base construction have created large and expressive ontologies that are often noisy and inconsistent, posing further challenges for conventional reasoners. To address these challenges, researchers have explored neuro-symbolic approaches that combine neural networks' learning capabilities with symbolic systems' reasoning abilities. In this chapter,we provide an overview of the existing literature in the field of neuro-symbolic deductive reasoning supported by RDF(S), the description logics EL and ALC, and OWL 2 RL, discussing the techniques employed, the tasks they address, and other relevant efforts in this area.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-palantir-digitalhealth-fdp-contract-redacted-2024", "title": "NHSE publishes heavily redacted Palantir FDP contract", "authors_or_org": "Digital Health", "year": 2024, "source_type": "news", "bucket": "palantir", "url": "https://www.digitalhealth.net/2024/01/nhse-publishes-heavily-redacted-palantir-fdp-contract", "doi_or_identifier": null, "venue_or_site": "Digital Health", "abstract_or_summary": "Trade-press reporting on NHS England publishing a heavily redacted FDP contract with Palantir, relevant to transparency and governance scrutiny.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "secondary_source", "retrieval_tags": [ "contract", "fdp", "independent-reporting", "nhs", "palantir", "transparency" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase2-garijo-2021-fair-semantic-artefacts", "title": "Nine best practices for research software registries and repositories: A concise guide", "authors_or_org": "Daniel Garijo; Oscar Corcho; et al.", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1371/journal.pcbi.1009333", "doi_or_identifier": "10.1371/journal.pcbi.1009333", "venue_or_site": "PLOS Computational Biology", "abstract_or_summary": "Concise best-practice guide for research software registries and repositories; useful for FAIR-style discoverability and governance of reusable digital artefacts including semantic tools.", "key_claims": [ "Research infrastructure benefits from metadata, identifiers, licensing, versioning, and community governance.", "Reusable software and semantic artefacts require curation practices beyond one-time publication.", "Ontology-backed AI knowledge bases should expose metadata, versioning, provenance, and reuse conditions." ], "ontology_relevance": "Supports FAIR-style management of semantic artefacts and ontology tooling in scientific infrastructure.", "ai_relevance": "Useful for making local AI knowledge bases reusable by other agents and researchers.", "palantir_relevance": "Useful contrast with proprietary platform governance and portability concerns.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "fair", "metadata", "provenance", "research-infrastructure", "reuse", "versioning" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-llm-ontmem-010", "title": "OAEI-LLM-T: A TBox Benchmark Dataset for Understanding LLM Hallucinations in Ontology Matching Systems", "authors_or_org": "Zhangcheng Qiang", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2503.21813", "doi_or_identifier": "arxiv:2503.21813", "venue_or_site": "arXiv", "abstract_or_summary": "Introduces an OAEI-derived TBox benchmark for studying and classifying hallucinations made by LLMs in ontology matching systems across schema-matching tasks.", "key_claims": [ "LLM hallucinations appear at the mapping and schema level, not only in natural-language answers.", "Ontology matching needs hallucination taxonomies tailored to TBox terminology and formal schema use.", "Benchmarking multiple LLMs on OAEI-derived tasks can support leaderboards and fine-tuning for ontology matching." ], "ontology_relevance": "Important benchmark source for reliability of LLM-based ontology alignment.", "ai_relevance": "Provides structured hallucination evaluation for a semantic interoperability task.", "palantir_relevance": "", "quality_signal": "preprint_benchmark", "retrieval_tags": [ "benchmark", "hallucination", "oaei", "ontology-matching", "tbox" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-std-oecd-ai-classification-framework-2022", "title": "OECD Framework for the Classification of AI Systems", "authors_or_org": "OECD", "year": 2022, "source_type": "technical_report", "bucket": "technical", "url": "https://doi.org/10.1787/cb6d9eca-en", "doi_or_identifier": "10.1787/cb6d9eca-en", "venue_or_site": "OECD Digital Economy Papers", "abstract_or_summary": "OECD framework for classifying AI systems by people and planet, economic context, data and input, AI model, task and output, and system lifecycle dimensions.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "ai-classification", "governance", "oecd", "policy", "risk-context", "system-taxonomy" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase7-ontobee-linked-ontology-server-2016", "title": "Ontobee: A linked ontology data server to support ontology term dereferencing, linkage, query and integration", "authors_or_org": "Jie Zheng; Zuoshuang Xiang; Jie Lin; et al.", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/nar/gkw918", "doi_or_identifier": "10.1093/nar/gkw918", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "Describes Ontobee, a linked ontology data server supporting term dereferencing, linkage, querying, and integration across biomedical ontologies.", "key_claims": [ "Ontology terms need dereferenceable web identifiers and linked-data access.", "Ontology servers can support SPARQL query, linkage, and integration across ontology resources.", "Linked ontology infrastructure improves reuse and machine access." ], "ontology_relevance": "Supports the semantic-web argument that ontology value depends on resolvable identifiers and service infrastructure.", "ai_relevance": "Linked ontology term services enable AI systems to dereference identifiers, retrieve term context, and connect domain semantics across sources.", "palantir_relevance": "Analogous to enterprise object and type registries exposed through APIs and governed query layers.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "linked-data", "ontobee", "ontology-repository", "phase7", "sparql", "term-dereferencing" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase11-tuan-sanyal-2026-ontology-constrained-neural-reasoning-enterprise-agents", "title": "Ontology-Constrained Neural Reasoning in Enterprise Agentic Systems: A Neurosymbolic Architecture for Domain-Grounded AI Agents", "authors_or_org": "Thanh Luong Tuan; Abhijit Sanyal", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2604.00555", "doi_or_identifier": "10.48550/arxiv.2604.00555", "venue_or_site": "arXiv", "abstract_or_summary": "Presents a neurosymbolic architecture for enterprise agentic systems that uses role, domain, and interaction ontologies to constrain LLM agents. The preprint reports cross-industry and cross-model experiments and argues for output-side validation, ontology-constrained tool discovery, compliance enforcement, and domain-grounded reasoning.", "key_claims": [ "Enterprise LLM adoption is constrained by hallucination, domain drift, and weak regulatory enforcement at the reasoning layer.", "A three-layer ontology model can ground roles, domains, and interactions for enterprise agents.", "The paper distinguishes input-side coupling from output-side ontological validation and compliance enforcement.", "Reported experiments suggest ontology-coupled agents can improve metric accuracy and role consistency, with larger value where parametric model knowledge is weaker.", "Enterprise LLM adoption is constrained by hallucination, domain drift, and difficulty enforcing compliance at the reasoning level.", "The paper proposes a three-layer ontology framework covering role, domain, and interaction ontologies for enterprise agents.", "The architecture distinguishes input-side constraints such as context assembly and tool discovery from output-side validation such as response checking and compliance enforcement.", "A controlled experiment across industries and LLMs reports improved metric accuracy and role consistency for ontology-coupled agents.", "The authors argue that ontology grounding is especially valuable where model parametric knowledge is weakest." ], "ontology_relevance": "Directly supports the research-agenda claim that agentic AI needs ontology-constrained reasoning and validation, while remaining preprint-level evidence.", "ai_relevance": "Highly relevant frontier preprint for ontology-constrained enterprise agents, neurosymbolic coupling, tool discovery, and reasoning validation.", "palantir_relevance": "Strong comparator for Palantir's ontology/AIP architecture because it also frames ontology as a domain-grounding and tool-discovery layer for enterprise agents.", "quality_signal": "preprint", "retrieval_tags": [ "agentic-ai", "domain-grounded-ai", "domain-ontology", "enterprise-agents", "governance", "governed-action", "interaction-ontology", "neurosymbolic", "neurosymbolic-ai", "ontology-constrained-reasoning", "phase11", "phase5", "role-ontology", "tool-discovery", "validation" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase6-openlineage-docs-specification", "title": "OpenLineage Documentation and Specification", "authors_or_org": "OpenLineage Project", "year": 2026, "source_type": "docs", "bucket": "technical", "url": "https://openlineage.io/docs", "doi_or_identifier": "openlineage specification", "venue_or_site": "OpenLineage", "abstract_or_summary": "Open standard for lineage metadata collection around runs, jobs, datasets, events, and extensible facets.", "key_claims": [ "Lineage should be emitted through shared event metadata rather than custom integrations.", "The core model identifies jobs, runs, and datasets consistently.", "Facets extend the model for system-specific metadata." ], "ontology_relevance": "Good operational lineage event model to connect pipelines, datasets, transformations, and provenance graphs.", "ai_relevance": "Can provide agents with impact analysis, freshness context, and explainability for data-derived answers.", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "data-lineage", "facets", "metadata-events", "openlineage", "phase6", "pipeline-observability", "provenance" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "ont-ai-012", "title": "OWL2Vec*: Embedding of OWL Ontologies", "authors_or_org": "Jiaoyan Chen; Pan Hu; Ernesto Jiménez-Ruiz; Ole Magnus Holter; Denvar Antonyrajah; Ian Horrocks", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://link.springer.com/article/10.1007/s10994-021-05997-6", "doi_or_identifier": "10.1007/s10994-021-05997-6", "venue_or_site": "Machine Learning", "abstract_or_summary": "Presents OWL2Vec*, a method for learning embeddings from OWL ontologies using graph walks, lexical information, and logical axioms.", "key_claims": [ "OWL ontologies can be embedded using structural, lexical, and logical signals.", "Ontology embeddings support downstream prediction and classification tasks.", "Combining semantics and graph context improves representation quality." ], "ontology_relevance": "Directly addresses vector representations of formal ontologies.", "ai_relevance": "Shows how OWL semantics can be integrated into machine learning workflows.", "palantir_relevance": "Relevant to semantic embeddings over enterprise ontology classes and instances.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "machine-learning", "ontology-embedding", "owl", "semantic-web" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-suchman-1987-plans-situated-actions", "title": "Plans and Situated Actions: The Problem of Human-Machine Communication", "authors_or_org": "Lucy A. Suchman", "year": 1987, "source_type": "book", "bucket": "books", "url": "https://doi.org/10.1017/CBO9780511625518", "doi_or_identifier": "10.1017/cbo9780511625518", "venue_or_site": "Cambridge University Press", "abstract_or_summary": "Classic HCI and STS book arguing that human action is situated and cannot be fully captured by pre-specified plans.", "key_claims": [ "Plans are resources for situated action, not complete determinants of behavior.", "Human-machine communication fails when systems assume context is fully specified in advance.", "Situated practice often exceeds formal representation." ], "ontology_relevance": "Important limitation source: no ontology can exhaustively encode the local context of action.", "ai_relevance": "Agent plans and tool schemas need runtime interpretation, human repair, and contextual judgment.", "palantir_relevance": "Useful critique of assuming object/action ontologies can fully specify operational reality.", "quality_signal": "widely_cited", "retrieval_tags": [ "context", "critique", "hci", "human-machine-communication", "phase3", "situated-action" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-gitelman-2013-raw-data-oxymoron", "title": "Raw Data Is an Oxymoron", "authors_or_org": "Lisa Gitelman, editor", "year": 2013, "source_type": "book", "bucket": "books", "url": "https://mitpress.mit.edu/9780262518284/raw-data-is-an-oxymoron", "doi_or_identifier": "isbn 9780262518284", "venue_or_site": "MIT Press", "abstract_or_summary": "Edited volume arguing that data are always already produced, formatted, selected, and interpreted through historical and technical practices.", "key_claims": [ "Data are not raw givens; they are made through instruments, formats, categories, and labor.", "What counts as data changes across historical and institutional settings.", "Data critique must examine the conditions of production and reuse." ], "ontology_relevance": "Important counterweight to ontology realism: ontology-backed data still reflect collection and categorization choices.", "ai_relevance": "AI grounding on data and graphs does not remove bias or interpretation if the underlying data production remains opaque.", "palantir_relevance": "Useful for discussing limitations of treating integrated enterprise data as an unmediated source of truth.", "quality_signal": "widely_cited", "retrieval_tags": [ "classification", "data-critique", "data-production", "epistemology", "phase3", "raw-data" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-std-unesco-ai-ram-2023", "title": "Readiness Assessment Methodology: A Tool of the Recommendation on the Ethics of Artificial Intelligence", "authors_or_org": "UNESCO", "year": 2023, "source_type": "assessment_methodology", "bucket": "technical", "url": "https://unesdoc.unesco.org/ark:/48223/pf0000385198", "doi_or_identifier": null, "venue_or_site": "UNESCO Methodology", "abstract_or_summary": "UNESCO methodology for assessing national readiness to implement ethical AI governance across legal, social, scientific, economic, cultural, technical, and institutional dimensions.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "ai-governance", "data-governance", "institutional-controls", "policy-capacity", "readiness-assessment", "unesco" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-std-oecd-ai-recommendation-2019", "title": "Recommendation of the Council on Artificial Intelligence", "authors_or_org": "OECD", "year": 2019, "source_type": "policy_instrument", "bucket": "technical", "url": "https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0449", "doi_or_identifier": null, "venue_or_site": "OECD Legal Instruments", "abstract_or_summary": "OECD legal instrument establishing principles for trustworthy AI and policy recommendations for national governments, including inclusive growth, human-centered values, transparency, robustness, accountability, investment, skills, and cooperation.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "accountability", "ai-principles", "governance", "oecd", "policy", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-std-eu-data-governance-act-2022", "title": "Regulation (EU) 2022/868 on European data governance (Data Governance Act)", "authors_or_org": "European Parliament, Council of the European Union", "year": 2022, "source_type": "regulation", "bucket": "technical", "url": "https://eur-lex.europa.eu/eli/reg/2022/868/oj", "doi_or_identifier": null, "venue_or_site": "Official Journal of the European Union", "abstract_or_summary": "EU regulation establishing governance mechanisms for data sharing, reuse of protected public-sector data, data intermediation services, data altruism, and European data innovation structures.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "data-governance-act", "data-intermediation", "data-sharing", "eu", "policy", "public-sector-data" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase7-relations-in-biomedical-ontologies-2005", "title": "Relations in biomedical ontologies", "authors_or_org": "Barry Smith; Werner Ceusters; Bert Klagges; et al.", "year": 2005, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1186/gb-2005-6-5-r46", "doi_or_identifier": "10.1186/gb-2005-6-5-r46", "venue_or_site": "Genome Biology", "abstract_or_summary": "Defines and motivates relations in biomedical ontologies, emphasizing the need for shared relation semantics such as part_of, develops_from, and located_in.", "key_claims": [ "Ontology relations need explicit semantics to support reliable reasoning.", "Shared relation vocabularies improve interoperability among biomedical ontologies.", "Poorly defined relations can cause incorrect graph inference." ], "ontology_relevance": "Foundational applied source for typed relations, relation governance, and cross-ontology interoperability.", "ai_relevance": "AI systems using graphs need relation semantics, not just nodes and labels, to avoid misleading inferences.", "palantir_relevance": "Directly relevant to Palantir-style link types: relation semantics determine what actions and reasoning are safe.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "interoperability", "ontology-relations", "phase7", "relation-semantics", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase11-soares-wassermann-2026-specific-domain-ontology-construction-llms", "title": "Specific Domain Ontology Construction Using Large Language Models", "authors_or_org": "Vivian Magri Alcaldi Soares; Renata Wassermann", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2606.20691", "doi_or_identifier": "10.48550/arxiv.2606.20691", "venue_or_site": "arXiv", "abstract_or_summary": "Experiments with LLMs acting as domain experts to construct conceptual hierarchies for the Brazilian maritime territory domain. Human expert evaluation found generally coherent conceptualizations but no output that was satisfactory without refinement.", "key_claims": [ "LLMs can generate generally coherent domain conceptual hierarchies for a specific domain.", "Human expert evaluation found that none of the generated ontologies was completely satisfactory without refinement.", "LLM-based ontology construction should be treated as assisted drafting rather than finished ontology engineering.", "The study reinforces the need for human-in-the-loop evaluation and domain-specific quality criteria." ], "ontology_relevance": "Useful cautionary evidence for LLM-assisted ontology engineering and expert validation.", "ai_relevance": "Current evidence that LLM-generated domain ontologies can be useful drafts but still require expert review and refinement.", "palantir_relevance": "Indirect comparator for enterprise ontology generation; it reinforces the need for expert governance behind automated modeling.", "quality_signal": "preprint", "retrieval_tags": [ "domain-ontology", "expert-evaluation", "human-in-the-loop", "llm", "ontology-construction", "ontology-engineering", "ontology-evaluation", "ontology-generation", "phase11" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-jasanoff-2004-states-knowledge", "title": "States of Knowledge: The Co-production of Science and Social Order", "authors_or_org": "Sheila Jasanoff, editor", "year": 2004, "source_type": "book", "bucket": "books", "url": "https://www.routledge.com/States-of-Knowledge-The-Co-production-of-Science-and-the-Social-Order/Jasanoff/p/book/9780415403290", "doi_or_identifier": "isbn 9780415403290", "venue_or_site": "Routledge", "abstract_or_summary": "Influential STS volume on co-production: knowledge systems and social orders are made together.", "key_claims": [ "Knowledge and social order are co-produced rather than separable domains.", "Institutions shape what counts as valid knowledge, and knowledge practices shape institutions.", "Governance controversies often turn on whose knowledge categories become authoritative." ], "ontology_relevance": "Frames ontology building as co-production of organizational knowledge and operational authority.", "ai_relevance": "AI governance must ask who defines valid data, labels, risk categories, and acceptable decisions.", "palantir_relevance": "Useful for analyzing ontology-backed platforms as tools that co-produce institutional visibility and action.", "quality_signal": "widely_cited", "retrieval_tags": [ "authority", "co-production", "institutions", "knowledge-governance", "phase3", "sts" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase4-std-eu-altai-2020", "title": "The Assessment List for Trustworthy Artificial Intelligence (ALTAI) for Self Assessment", "authors_or_org": "High-Level Expert Group on Artificial Intelligence, European Commission", "year": 2020, "source_type": "assessment_tool", "bucket": "technical", "url": "https://digital-strategy.ec.europa.eu/en/library/assessment-list-trustworthy-artificial-intelligence-altai-self-assessment", "doi_or_identifier": null, "venue_or_site": "European Commission Assessment Tool", "abstract_or_summary": "Self-assessment checklist operationalizing the EU trustworthy AI guidelines across human oversight, robustness, privacy, transparency, diversity, societal wellbeing, and accountability.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "altai", "checklist", "eu", "evaluation", "governance-artifacts", "self-assessment", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-kitchin-2014-data-revolution", "title": "The Data Revolution: Big Data, Open Data, Data Infrastructures and Their Consequences", "authors_or_org": "Rob Kitchin", "year": 2014, "source_type": "book", "bucket": "books", "url": "https://uk.sagepub.com/en-gb/eur/the-data-revolution/book242119", "doi_or_identifier": "isbn 9781446287484", "venue_or_site": "SAGE", "abstract_or_summary": "Critical social-science account of big data, data infrastructures, analytics, governance, and social consequences.", "key_claims": [ "Data are produced through socio-technical systems, not simply collected from the world.", "Data infrastructures embed choices about measurement, access, ownership, and analysis.", "Big-data systems create new governance, surveillance, and epistemic risks." ], "ontology_relevance": "Helps frame ontology as part of data infrastructure, with consequences for what becomes measurable and governable.", "ai_relevance": "AI systems depend on data infrastructures whose production conditions shape model behavior and risk.", "palantir_relevance": "Useful background for analyzing platforms that make institutional data analytics-ready and AI-ready.", "quality_signal": "widely_cited", "retrieval_tags": [ "big-data", "critique", "data-governance", "data-infrastructure", "phase3", "surveillance" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-star-1999-ethnography-infrastructure", "title": "The Ethnography of Infrastructure", "authors_or_org": "Susan Leigh Star", "year": 1999, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1177/00027649921955326", "doi_or_identifier": "10.1177/00027649921955326", "venue_or_site": "American Behavioral Scientist", "abstract_or_summary": "Short influential article defining infrastructure relationally and arguing for ethnographic attention to standards, embeddedness, and breakdown.", "key_claims": [ "Infrastructure is relational: what is background for one group can be foreground work for another.", "Infrastructure is embedded in other structures, learned as membership, and visible upon breakdown.", "Standards and installed bases shape what future systems can do." ], "ontology_relevance": "Frames ontology as infrastructure whose maintenance and politics become visible at points of mismatch and failure.", "ai_relevance": "AI failures often expose hidden data, schema, and workflow infrastructure beneath model outputs.", "palantir_relevance": "Useful for article sections on ontology maintenance, migration, and operational breakdowns.", "quality_signal": "widely_cited", "retrieval_tags": [ "breakdown", "ethnography", "infrastructure", "installed-base", "phase3", "standards" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "n3c-governance-ecosystem-2022", "title": "The N3C governance ecosystem: A model socio-technical partnership for collaborative analytics at scale", "authors_or_org": "N3C Consortium authors", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8822626", "doi_or_identifier": null, "venue_or_site": "Journal of Clinical and Translational Science / PubMed Central", "abstract_or_summary": "Peer-reviewed paper describing N3C's governance ecosystem, including socio-technical partnership, access controls, data-use processes, and collaborative analytics at scale.", "key_claims": [ "Large collaborative data enclaves require governance ecosystems, not only technical infrastructure.", "N3C governance includes access processes, data-use agreements, committees, and operational coordination.", "The paper is a useful comparator for public-sector governance around Palantir-style data platforms." ], "ontology_relevance": "Supports analysis of ontology/data platforms as socio-technical systems with committees, policies, and access workflows.", "ai_relevance": "AI-ready clinical data environments require governance mechanisms before model development and deployment.", "palantir_relevance": "Provides governance comparator for Palantir-associated health-data infrastructure without relying on vendor marketing.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "clinical-data", "collaborative-analytics", "governance", "health-ai", "n3c", "socio-technical" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-n3c-governance-ecosystem-2023", "title": "The N3C governance ecosystem: A model socio-technical partnership for large-scale health data sharing", "authors_or_org": "N3C authors", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://pmc.ncbi.nlm.nih.gov/articles/PMC10789985", "doi_or_identifier": null, "venue_or_site": "PMC / peer-reviewed article", "abstract_or_summary": "Peer-reviewed article presenting N3C as a public-private-government partnership and describing governance as a socio-technical ecosystem around a centralized data enclave.", "key_claims": [ "N3C is presented as a public-private-government partnership around a centralized data enclave.", "Governance is treated as a socio-technical ecosystem, not merely access control software.", "Large-scale health data sharing requires committees, agreements, norms, and technical infrastructure." ], "ontology_relevance": "Useful counterweight to purely technical ontology/platform narratives.", "ai_relevance": "AI deployment on health data needs institutional governance and social legitimacy.", "palantir_relevance": "Indirect comparator for Palantir public-health deployments and governance expectations.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "data-enclave", "governance", "health-data-sharing", "n3c", "socio-technical" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase3-parsons-wand-2008-ontology-conceptual-modeling", "title": "Using Cognitive Principles to Guide Classification in Information Systems Modeling", "authors_or_org": "Jeffrey Parsons and Yair Wand", "year": 2008, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1287/mnsc.1070.0814", "doi_or_identifier": "10.1287/mnsc.1070.0814", "venue_or_site": "Management Science", "abstract_or_summary": "Paper connecting classification, cognition, and conceptual modeling in information systems.", "key_claims": [ "Classification choices in information systems should account for how users understand domain categories.", "A model can be formally precise but cognitively misaligned with stakeholders.", "Conceptual modeling quality includes human comprehension as well as formal adequacy." ], "ontology_relevance": "Adds a human-cognitive dimension to ontology design and classification quality.", "ai_relevance": "LLM and agent interfaces depend on models that humans can inspect, query, and correct.", "palantir_relevance": "Relevant to enterprise ontology adoption where business users must recognize and trust object categories.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "classification", "cognition", "conceptual-modeling", "phase3", "sociotechnical", "user-understanding" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "phase6-xiang-2025-when-to-use-graphs-rag", "title": "When to Use Graphs in RAG: A Comprehensive Analysis for Graph Retrieval-Augmented Generation", "authors_or_org": "Zhishang Xiang; Chuanjie Wu; Qinggang Zhang; Shengyuan Chen; Zijin Hong; Xiao Huang; Jinsong Su", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2506.05690", "doi_or_identifier": "10.48550/arxiv.2506.05690", "venue_or_site": "arXiv / GraphRAG-Bench", "abstract_or_summary": "Introduces GraphRAG-Bench to test when graph structures help RAG across fact retrieval, reasoning, summarization, and generation tasks.", "key_claims": [ "GraphRAG does not always outperform vanilla RAG.", "Graph structure helps most under hierarchical retrieval and deeper contextual reasoning needs.", "Evaluation should cover graph construction, retrieval, and generation end to end." ], "ontology_relevance": "Useful for deciding when ontology/KG structure is worth adding to retrieval systems.", "ai_relevance": "Directly addresses GraphRAG evaluation and LLM retrieval design tradeoffs.", "palantir_relevance": "", "quality_signal": "preprint_benchmark", "retrieval_tags": [ "benchmark", "evaluation", "graphrag", "graphrag-bench", "hierarchical-retrieval", "phase6", "rag" ], "triage_tier": "candidate", "triage_score": 48 }, { "id": "oa-https-doi-org-10-1109-access-2019-2953499", "title": "A Survey on Digital Twin: Definitions, Characteristics, Applications, and Design Implications", "authors_or_org": "Barbara Rita Barricelli, Elena Casiraghi, Daniela Fogli", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://ieeexplore.ieee.org/document/8901113", "doi_or_identifier": "10.1109/access.2019.2953499", "venue_or_site": "IEEE Access", "abstract_or_summary": "When, in 1956, Artificial Intelligence (AI) was officially declared a research field, no one would have ever predicted the huge influence and impact its description, prediction, and prescription capabilities were going to have on our daily lives. In parallel to continuous advances in AI, the past decade has seen the spread of broadband and ubiquitous connectivity, (embedded) sensors collecting descriptive high dimensional data, and improvements in big data processing techniques and cloud computing. The joint usage of such technologies has led to the creation of digital twins, artificial intelligent virtual replicas of physical systems. Digital Twin (DT) technology is nowadays being developed and commercialized to optimize several manufacturing and aviation processes, while in the healthcare and medicine fields this technology is still at its early development stage. This paper presents the results of a study focused on the analysis of the state-of-the-art definitions of DT, the investigation of the main characteristics that a DT should possess, and the exploration of the domains in which DT applications are currently being developed. The design implications derived from the study are then presented: they focus on socio-technical design aspects and DT lifecycle. Open issues and challenges that require to be addressed in the future are finally discussed.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "applications", "commercial", "definitions", "digital-twin", "enterprise-modeling", "openalex", "socio-technical" ], "triage_tier": "candidate", "triage_score": 47 }, { "id": "oa-https-doi-org-10-1016-j-inffus-2023-101896", "title": "Connecting the dots in trustworthy Artificial Intelligence: From AI principles, ethics, and key requirements to responsible AI systems and regulation", "authors_or_org": "Natalia Díaz-Rodríguez, Javier Del Ser, Mark Coeckelbergh, Marcos López de Prado, Enrique Herrera‐Viedma, Francisco Herrera", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://openalex.org/W4381848566", "doi_or_identifier": "10.1016/j.inffus.2023.101896", "venue_or_site": "Information Fusion", "abstract_or_summary": "Trustworthy Artificial Intelligence (AI) is based on seven technical requirements sustained over three main pillars that should be met throughout the system’s entire life cycle: it should be (1) lawful, (2) ethical, and (3) robust, both from a technical and a social perspective. However, attaining truly trustworthy AI concerns a wider vision that comprises the trustworthiness of all processes and actors that are part of the system’s life cycle, and considers previous aspects from different lenses. A more holistic vision contemplates four essential axes: the global principles for ethical use and development of AI-based systems, a philosophical take on AI ethics, a risk-based approach to AI regulation, and the mentioned pillars and requirements. The seven requirements (human agency and oversight; robustness and safety; privacy and data governance; transparency; diversity, non-discrimination and fairness; societal and environmental wellbeing; and accountability) are analyzed from a triple perspective: What each requirement for trustworthy AI is, Why it is needed, and How each requirement can be implemented in practice. On the other hand, a practical approach to implement trustworthy AI systems allows defining the concept of responsibility of AI-based systems facing the law, through a given auditing process. Therefore, a responsible AI system is the resulting notion we introduce in this work, and a concept of utmost necessity that can be realized through auditing processes, subject to the challenges posed by the use of regulatory sandboxes. Our multidisciplinary vision of trustworthy AI culminates in a debate on the diverging views published lately about the future of AI. Our reflections in this matter conclude that regulation is a key for reaching a consensus among these views, and that trustworthy and responsible AI systems will be crucial for the present and future of our society.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "commercial", "openalex" ], "triage_tier": "candidate", "triage_score": 47 }, { "id": "s2-10-1109-ainit65432-2025-11035727", "title": "AI-Based Intelligent Decision-Making Platform for Operational Excellence in the Oil and Gas Industry", "authors_or_org": "Xiaoxiong Li, Shuanglei Peng, Xuejia Ma, Zhizhao Tian, Xiao He, Yin Wang", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/82c872b115684e0d8d73b8b687f4c01ec50b9126", "doi_or_identifier": "10.1109/ainit65432.2025.11035727", "venue_or_site": "2025 IEEE 6th International Seminar on Artificial Intelligence, Networking and Information Technology (AINIT)", "abstract_or_summary": "The integration of artificial intelligence (AI) in the oil and gas industry has revolutionized operational processes, enabling organizations to overcome complex challenges and improve efficiency. This study has explored the development and implementation of a suite of AI-based intelligent systems by CNPCNP Petroleum S.A. (CNPCNP) to enhance decision-making, optimize resource utilization, and address critical challenges in its Niger operations. By leveraging advanced AI technologies, including deep learning, business intelligence (BI), natural language processing (NLP), and robotic process automation (RPA), CNPCNP has developed the Intelligent Decision-Making Platform for finance, planning, human resources, security, and health management. The Intelligent Decision-Making Platform has provided an integrated governance and data intelligence system, connecting all the “silo” data and creating infinite intelligent analytical data models with very limited amount of code and “drag-and-drop” methods. This system offers real-time insights, streamline operations, and improve organizational performance. This paper serves as a reference for leveraging AI in resourcelimited and high-security environments, contributing to the broader discourse on digital transformation in the energy industry.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "oa-https-doi-org-10-1057-s42984-023-00068-7", "title": "Algorithmic predictions and pre-emptive violence: artificial intelligence and the future of unmanned aerial systems", "authors_or_org": "Anthony Downey", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4389334393", "doi_or_identifier": "10.1057/s42984-023-00068-7", "venue_or_site": "Digital War", "abstract_or_summary": "Abstract The military rationale of a pre-emptive strike is predicated upon the calculation and anticipation of threat. The underlying principle of anticipation, or prediction, is foundational to the operative logic of AI. The deployment of predictive, algorithmically driven systems in unmanned aerial systems (UAS) would therefore appear to be all but inevitable. However, the fatal interlocking of martial paradigms of pre-emption and models of predictive analysis needs to be questioned insofar as the irreparable decisiveness of a pre-emptive military strike is often at odds with the probabilistic predictions of AI. The pursuit of a human right to protect communities from aerial threats needs to therefore consider the degree to which algorithmic auguries—often erroneous but nevertheless evident in the prophetic mechanisms that power autonomous aerial apparatuses—essentially authorise and further galvanise the long-standing martial strategy of pre-emption. In the context of unmanned aerial systems, this essay will outline how AI actualises and summons forth “threats” through (i) the propositional logic of algorithms (their inclination to yield actionable directives); (ii) the systematic training of neural networks (through habitually biased methods of data-labelling); and (iii) a systemic reliance on models of statistical analysis in the structural design of machine learning (which can and do produce so-called “hallucinations”). Through defining the deterministic intentionality, systematic biases and systemic dysfunction of algorithms, I will identify how individuals and communities—configured upon and erroneously flagged through the machinations of so-called “black box” instruments—are invariably exposed to the uncertainty (or brute certainty) of imminent death based on algorithmic projections of “threat”.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase3-privacy-intl-palantir-2021", "title": "All roads lead to Palantir?", "authors_or_org": "Privacy International", "year": 2021, "source_type": "webpage", "bucket": "synthesis", "url": "https://privacyinternational.org/report/4271/all-roads-lead-palantir", "doi_or_identifier": null, "venue_or_site": "Privacy International", "abstract_or_summary": "Privacy International report mapping Palantir's public-sector and crisis-response footprint and raising rights-based concerns about surveillance, procurement, opacity, and state analytics infrastructures.", "key_claims": [ "Privacy International frames Palantir as a significant actor in public-sector data analytics infrastructures.", "The report raises concerns about transparency, surveillance, procurement, and accountability.", "It is rights-based NGO evidence, useful for critique but not for vendor technical mechanics." ], "ontology_relevance": "Connects ontology-like data integration to institutional power and visibility.", "ai_relevance": "Agentic AI over public-sector data intensifies existing surveillance and accountability concerns.", "palantir_relevance": "Broad civil-liberties critique of Palantir's government-facing business.", "quality_signal": "secondary_source", "retrieval_tags": [ "accountability", "civil-liberties", "palantir", "privacy-international", "public-sector", "surveillance" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "munn_smith_2008_applied_ontology", "title": "Applied Ontology: An Introduction", "authors_or_org": "Katherine Munn and Barry Smith, editors", "year": 2008, "source_type": "book", "bucket": "books", "url": "https://www.degruyterbrill.com/document/doi/10.1515/9783110324860/html", "doi_or_identifier": "10.1515/9783110324860; isbn 9783110324860", "venue_or_site": "ontos / De Gruyter", "abstract_or_summary": "Introductory volume connecting philosophical category analysis with practical ontology development for information systems, science, government, industry, and healthcare.", "key_claims": [ "Applied ontology brings philosophical category analysis into information management.", "Interoperability problems often arise from incompatible or implicit category systems.", "Ontology engineering requires attention to reality, representation, and practical use." ], "ontology_relevance": "Accessible bridge from philosophical ontology to applied information-system ontology.", "ai_relevance": "Supports the argument that AI knowledge infrastructure needs category discipline and semantic interoperability.", "palantir_relevance": "Relevant to enterprise object modeling, operational semantics, and cross-system integration.", "quality_signal": "primary_source", "retrieval_tags": [ "applied-ontology", "category-analysis", "interoperability", "philosophy" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-dcmi-terms-2020", "title": "DCMI Metadata Terms", "authors_or_org": "Dublin Core Metadata Initiative Usage Board", "year": 2020, "source_type": "standard", "bucket": "technical", "url": "https://www.dublincore.org/specifications/dublin-core/dcmi-terms", "doi_or_identifier": null, "venue_or_site": "Dublin Core Metadata Initiative Recommendation", "abstract_or_summary": "Defines DCMI properties, classes, vocabulary encoding schemes, and syntax encoding schemes for describing resources, provenance-adjacent metadata, rights, formats, dates, and relations.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "cataloging", "dcmi", "linked-data", "metadata", "resource-description", "semantic-interoperability" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "industry-commercial-enterprise-knowledge-critique", "title": "Enterprise Knowledge practitioner articles on knowledge graph strategy and pitfalls", "authors_or_org": "Enterprise Knowledge", "year": 2026, "source_type": "technical_article", "bucket": "commercial", "url": "https://enterprise-knowledge.com", "doi_or_identifier": null, "venue_or_site": "Enterprise Knowledge Blog", "abstract_or_summary": "Enterprise Knowledge practitioner writing discusses KG strategy, ontology governance, taxonomy design, semantic layers, and common failure modes in enterprise knowledge initiatives.", "key_claims": [ "Successful KG work should start with bounded use cases and competency questions.", "Governance, ownership, and maintenance are frequent failure points.", "Taxonomy, ontology, and data modeling choices should match the business problem rather than technology fashion." ], "ontology_relevance": "Provides pragmatic critique and implementation guidance from ontology/KG consultants.", "ai_relevance": "Highlights governance and semantic quality issues that affect AI retrieval and grounding.", "palantir_relevance": "Useful external critique for judging ontology platform adoption beyond vendor hype.", "quality_signal": "secondary_source", "retrieval_tags": [ "consulting", "critique", "enterprise-knowledge", "governance", "ontology-strategy", "pitfalls" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase3-llm-ontmem-014", "title": "From Matching to Retrieval: A New Role for LLMs in Ontology Alignment", "authors_or_org": "Ontology Matching 2025 authors", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://ceur-ws.org/Vol-4144/om2025-LTpaper1.pdf", "doi_or_identifier": "ceur-ws vol-4144 om2025-ltpaper1", "venue_or_site": "Ontology Matching Workshop 2025 / CEUR-WS", "abstract_or_summary": "Explores a retrieval-oriented role for LLMs in ontology alignment, emphasizing semantically enriched candidate representations and refinement strategies for robust matching pipelines.", "key_claims": [ "LLMs can contribute to ontology alignment by enriching candidate representations, not only by directly deciding matches.", "Retrieval and refinement are important stages for robust LLM-based alignment pipelines.", "Precision must be balanced against probabilistic confidence when LLMs participate in matching." ], "ontology_relevance": "Adds a retrieval-centered perspective to ontology alignment, useful for large ontology matching pipelines.", "ai_relevance": "Shows LLMs as components in candidate retrieval and representation enrichment rather than black-box matchers.", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "candidate-generation", "llm", "om-2025", "ontology-alignment", "retrieval" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-iso-iec-22989-2022", "title": "ISO/IEC 22989:2022 Information technology - Artificial intelligence - Artificial intelligence concepts and terminology", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2022, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/74296.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for the AI concepts and terminology standard used to align communications among stakeholders and support development of related AI standards.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "ai-terminology", "governance", "iec", "iso", "semantic-interoperability", "standard-vocabulary" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-iso-iec-23053-2022", "title": "ISO/IEC 23053:2022 Framework for Artificial Intelligence (AI) Systems Using Machine Learning (ML)", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2022, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/74438.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for a conceptual framework and shared terminology describing components and functions of ML-based AI systems.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "ai-system-framework", "architecture", "governance", "iec", "iso", "machine-learning", "terminology" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-iso-iec-23894-2023", "title": "ISO/IEC 23894:2023 Information technology - Artificial intelligence - Guidance on risk management", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2023, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/77304.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for guidance on managing AI-related risks and integrating AI risk management into organizational activities and functions.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "ai-risk-management", "controls", "governance", "iec", "iso", "lifecycle", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-iso-iec-38507-2022", "title": "ISO/IEC 38507:2022 Information technology - Governance of IT - Governance implications of the use of artificial intelligence by organizations", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2022, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/56641.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for guidance to governing bodies, executives, auditors, policymakers, and service providers on governing current and future organizational uses of AI.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "ai-governance", "audit", "board-governance", "iec", "iso", "organizational-controls", "policy" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-iso-iec-5259-2-2024", "title": "ISO/IEC 5259-2:2024 Artificial intelligence - Data quality for analytics and machine learning (ML) - Part 2: Data quality measures", "authors_or_org": "ISO/IEC JTC 1/SC 42", "year": 2024, "source_type": "standard_metadata", "bucket": "technical", "url": "https://www.iso.org/standard/81860.html", "doi_or_identifier": null, "venue_or_site": "ISO/IEC International Standard", "abstract_or_summary": "Public ISO metadata for data quality models and measurable characteristics used to assess and report data quality for analytics and machine learning.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "data-quality-measures", "evaluation", "governance", "iec", "iso", "machine-learning", "metrics" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-json-ld-11-2020", "title": "JSON-LD 1.1: A JSON-based Serialization for Linked Data", "authors_or_org": "Gregg Kellogg, Pierre-Antoine Champin, Dave Longley, W3C JSON-LD Working Group", "year": 2020, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/json-ld11", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines JSON-LD 1.1 syntax and processing concepts for expressing linked data in JSON, including contexts, IRIs, nodes, graphs, framing, and compaction/expansion patterns.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "api-interoperability", "json-ld", "knowledge-graph", "linked-data", "semantic-web", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "industry-commercial-merck-cake-neptune", "title": "Merck uses Amazon Neptune for Change Assessment Knowledge Engine", "authors_or_org": "Amazon Web Services / Merck", "year": 2021, "source_type": "case_study", "bucket": "commercial", "url": "https://aws.amazon.com/solutions/case-studies/merck", "doi_or_identifier": null, "venue_or_site": "AWS Case Studies", "abstract_or_summary": "AWS case material describes Merck's Change Assessment Knowledge Engine, a graph-based system for assessing manufacturing process changes and related dependencies.", "key_claims": [ "Change impact analysis can be modeled as traversal over connected process, product, material, and compliance entities.", "Graph relationships make upstream and downstream dependencies more inspectable.", "Knowledge graphs can support regulated manufacturing decision workflows." ], "ontology_relevance": "Concrete example of domain ontology and dependency graph supporting operational decisions.", "ai_relevance": "The same dependency graph can supply structured context for AI-assisted change assessment.", "palantir_relevance": "Strong non-Palantir example of ontology-like operational decision support in industry.", "quality_signal": "primary_source", "retrieval_tags": [ "aws", "case-study", "impact-analysis", "manufacturing", "merck", "neptune" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-csvw-metadata-2015", "title": "Metadata Vocabulary for Tabular Data", "authors_or_org": "Jeni Tennison, Gregg Kellogg, Ivan Herman, W3C CSV on the Web Working Group", "year": 2015, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/csvw-metadata", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Specifies a metadata vocabulary for describing tables, schemas, columns, datatypes, foreign keys, transformations, and annotations for tabular data on the Web.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "csvw", "data-integration", "metadata", "schema", "semantic-interoperability", "tabular-data", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-odrl-vocab-2018", "title": "ODRL Vocabulary and Expression 2.2", "authors_or_org": "Renato Iannella, Serena Villata, W3C Permissions and Obligations Expression Working Group", "year": 2018, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/odrl-vocab", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Companion vocabulary and encodings for ODRL policies, including RDF terms for actions, constraints, logical operands, duties, remedies, and policy metadata.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "odrl", "policy-vocabulary", "rdf", "semantic-interoperability", "usage-control", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "oa-https-doi-org-10-3390-s23031658", "title": "PALANTIR: An NFV-Based Security-as-a-Service Approach for Automating Threat Mitigation", "authors_or_org": "Maxime Compastié, Antonio López Martínez, Carolina Fernández, Manuel Gil Pérez, Stylianos Tsarsitalidis, George Xylouris, Izidor Mlakar, Michail‐Alexandros Kourtis", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4319160095", "doi_or_identifier": "10.3390/s23031658", "venue_or_site": "Sensors", "abstract_or_summary": "Small and medium enterprises are significantly hampered by cyber-threats as they have inherently limited skills and financial capacities to anticipate, prevent, and handle security incidents. The EU-funded PALANTIR project aims at facilitating the outsourcing of the security supervision to external providers to relieve SMEs/MEs from this burden. However, good practices for the operation of SME/ME assets involve avoiding their exposure to external parties, which requires a tightly defined and timely enforced security policy when resources span across the cloud continuum and need interactions. This paper proposes an innovative architecture extending Network Function Virtualisation to externalise and automate threat mitigation and remediation in cloud, edge, and on-premises environments. Our contributions include an ontology for the decision-making process, a Fault-and-Breach-Management-based remediation policy model, a framework conducting remediation actions, and a set of deployment models adapted to the constraints of cloud, edge, and on-premises environment(s). Finally, we also detail an implementation prototype of the framework serving as evaluation material.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "s2-10-1038-s41598-025-29732-6", "title": "The construction of an integrated cloud network digital intelligence platform for rail transit based on artificial intelligence", "authors_or_org": "Keke Wang, Xin Zhou, J. Guan", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/3bf1bbc3d87b31f7add487fed0da021648a09c54", "doi_or_identifier": "10.1038/s41598-025-29732-6", "venue_or_site": "Scientific Reports", "abstract_or_summary": "This study presents the design and validation of a closed-loop control platform for rail transit construction. The platform integrates multi-source data, enables real-time prediction, and supports AI-driven scheduling, with strategy execution and feedback implemented via digital twins. A three-layer architecture is constructed, comprising edge sensing, cloud computing, and intelligent interaction. The system incorporates data fusion middleware, an AI decision engine, and a 3D digital twins module. The operational workflow follows the perception–fusion–prediction/optimization–execution/feedback loop: edge devices collect on-site status, cloud middleware integrates and serves the data, the AI engine performs prediction and scheduling optimization, and the digital twins layer validates strategies and dispatches execution to the front end. At the data modeling level, a Transformer-Encoder-based multimodal temporal fusion model is designed, and graph attention networks are employed for heterogeneous structure modeling. Apache Kafka and Flink handle streaming data to achieve high-frequency, low-latency processing. The intelligent analysis layer integrates a Spatio-Temporal Graph Convolutional Network for passenger flow and construction period prediction, a Shifted Window Transformer for image recognition, and the Proximal Policy Optimization (PPO) algorithm for task scheduling optimization. Field tests in an urban rail construction project show that the platform maintains 91.6% accuracy in passenger flow prediction under high-concurrency conditions and achieves 98.2% accuracy in image recognition. PPO-based scheduling reduces average task completion time by 27.4%. The system sustains an average response latency of 280 ms, peak throughput of 27,000 messages per second, and over 95% closed-loop execution success rate. These results indicate that the platform meets its design targets in prediction accuracy, response latency, and scheduling efficiency under real-world conditions, providing a foundation for informatization and intelligent upgrading in urban rail transit.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-rdf-data-cube-2014", "title": "The RDF Data Cube Vocabulary", "authors_or_org": "Richard Cyganiak, Dave Reynolds, W3C Government Linked Data Working Group", "year": 2014, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/vocab-data-cube", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines an RDF vocabulary for publishing multi-dimensional statistical data and observations with dimensions, measures, attributes, datasets, slices, and data structure definitions.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "data-cube", "metrics", "observations", "rdf", "semantic-interoperability", "statistics", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-vc-data-model-2-2025", "title": "Verifiable Credentials Data Model v2.0", "authors_or_org": "Manu Sporny, Dave Longley, David Chadwick, Ivan Herman, W3C Verifiable Credentials Working Group", "year": 2025, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/vc-data-model-2.0", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines an extensible data model for tamper-evident credentials, presentations, claims, issuers, holders, verifiers, credential status, schemas, evidence, and terms of use.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "attestation", "claims", "governance-artifacts", "identity", "trust", "verifiable-credentials", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase4-std-w3c-annotation-model-2017", "title": "Web Annotation Data Model", "authors_or_org": "Robert Sanderson, Paolo Ciccarese, Benjamin Young, W3C Web Annotation Working Group", "year": 2017, "source_type": "standard", "bucket": "technical", "url": "https://www.w3.org/TR/annotation-model", "doi_or_identifier": null, "venue_or_site": "W3C Recommendation", "abstract_or_summary": "Defines a structured model for annotations that associate bodies and targets, including motivations, selectors, states, agents, and provenance metadata.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_standard", "retrieval_tags": [ "annotation", "dataset-documentation", "governance-artifacts", "human-feedback", "provenance", "w3c" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "phase3-good-law-project-nhs-palantir-privacy-2023", "title": "Why we're working to uphold the privacy of NHS patient data", "authors_or_org": "Good Law Project", "year": 2023, "source_type": "webpage", "bucket": "synthesis", "url": "https://goodlawproject.org/update/why-were-working-to-uphold-the-privacy-of-nhs-patient-data", "doi_or_identifier": null, "venue_or_site": "Good Law Project", "abstract_or_summary": "Campaign update explaining legal/privacy concerns about Palantir's NHS data role, while noting NHS England's position about confidentiality and directed access.", "key_claims": [ "Good Law Project argues that lack of scrutiny and safeguards could lead to misuse of NHS data.", "The page acknowledges NHS England's claim that data access is governed by contract and NHS direction.", "It is legal advocacy evidence focused on safeguards and transparency." ], "ontology_relevance": "Shows how legal accountability questions attach to platform-mediated operational data use.", "ai_relevance": "AI governance requires enforceable safeguards and review mechanisms, not only internal controls.", "palantir_relevance": "Direct civil-society/legal challenge context for Palantir NHS work.", "quality_signal": "secondary_source", "retrieval_tags": [ "good-law-project", "legal-challenge", "nhs", "palantir", "privacy", "safeguards" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "masolo_2003_wonderweb_dolce", "title": "WonderWeb Deliverable D18: Ontology Library", "authors_or_org": "Claudio Masolo, Stefano Borgo, Aldo Gangemi, Nicola Guarino, Alessandro Oltramari, and Luc Schneider", "year": 2003, "source_type": "technical_article", "bucket": "technical", "url": "https://www.loa.istc.cnr.it/old/Papers/D18.pdf", "doi_or_identifier": "wonderweb deliverable d18", "venue_or_site": "WonderWeb Project / Laboratory for Applied Ontology", "abstract_or_summary": "Presents the WonderWeb foundational ontology library, including DOLCE, with emphasis on descriptive, cognitive, and linguistic categories.", "key_claims": [ "DOLCE is a descriptive ontology designed around cognitive and linguistic categories.", "Foundational ontology libraries can support domain ontology interoperability.", "Events, qualities, social objects, and descriptions require careful upper-level modeling." ], "ontology_relevance": "Primary source for DOLCE and the WonderWeb foundational ontology library.", "ai_relevance": "Relevant to natural-language-facing AI systems that must model descriptions, roles, events, and social categories.", "palantir_relevance": "Useful lens for workflows, organizational objects, plans, and decision records in enterprise ontology.", "quality_signal": "primary_source", "retrieval_tags": [ "dolce", "events", "foundational-ontology", "social-objects", "wonderweb" ], "triage_tier": "candidate", "triage_score": 46 }, { "id": "oa-https-doi-org-10-4230-tgdk-1-1-7", "title": "Embedding Entities and Relations for Learning and Inference in Knowledge Bases", "authors_or_org": "Yang, Bishan, Wen-tau Yih, He, Xiaodong, Gao, Jianfeng, Deng, Li", "year": 2014, "source_type": "paper", "bucket": "academic", "url": "https://openalex.org/W1533230146", "doi_or_identifier": "10.4230/tgdk.1.1.7", "venue_or_site": "arXiv (Cornell University)", "abstract_or_summary": "Compared to black-box neural networks, logic rules express explicit knowledge, can provide human-understandable explanations for reasoning processes, and have found their wide application in knowledge graphs and other downstream tasks. As extracting rules manually from large knowledge graphs is labour-intensive and often infeasible, automated rule learning has recently attracted significant interest, and a number of approaches to rule learning for knowledge graphs have been proposed. This survey aims to provide a review of approaches and a classification of state-of-the-art systems for learning first-order logic rules over knowledge graphs. A comparative analysis of various approaches to rule learning is conducted based on rule language biases, underlying methods, and evaluation metrics. The approaches we consider include inductive logic programming (ILP)-based, statistical path generalisation, and neuro-symbolic methods. Moreover, we highlight important and promising application scenarios of rule learning, such as rule-based knowledge graph completion, fact checking, and applications in other research areas.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "ontology_ai", "openalex" ], "triage_tier": "candidate", "triage_score": 45 }, { "id": "s2-10-1109-compsac61105-2024-00054", "title": "An Ontology Alignment Validation Approach Based on Supervised Machine Learning Algorithms and Automatic Schema Matching Approach", "authors_or_org": "Faten Abbassi, Y. Hlaoui", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://www.semanticscholar.org/paper/f2a2a72930b8cc4992bd3b186c1701cbe35ca728", "doi_or_identifier": "10.1109/compsac61105.2024.00054", "venue_or_site": "Annual International Computer Software and Applications Conference", "abstract_or_summary": "The existence of various representations of the same ontology poses a challenge in terms of manipulating knowledge across different computational domains. To address this issue, it would be prudent to harmonise similar ontologies by reducing their level of heterogeneity. This proposed solution involves aligning comparable ontologies through the utilisation of established ontology schema-matching techniques. In this paper, we introduce an approach for ontology alignment that leverages these techniques along with machine learning algorithms. To accomplish this, we propose a method for constructing a matrix based on ontology matching techniques, specifically employing element matching and structure matching techniques facilitated by elementary matchers. Subsequently, once the matrix is constructed, we use a composite matcher as a classifier to determine the degree of similarity between the two ontologies. Since these matchers are not readily available, we put forth the idea of implementing them in this paper using various supervised machine learning algorithms such as Neural Network and Logistic Regression. To validate our approach, we conducted an empirical assessment using the Conference track and the benchmark track of the reference ontologies provided by the Ontology Alignment Evaluation Initiative (OAEI 1).", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "foundational", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 44 }, { "id": "s2-10-52710-cfs-1008", "title": "Generative AI–Driven Semantic Integration Architecture for SAP Cloud and Hybrid Landscapes", "authors_or_org": "Umamaheswarareddy Chintam", "year": 2026, "source_type": "paper", "bucket": "academic", "url": "https://www.semanticscholar.org/paper/138ab87d1b4ba7bcef5a97d4d676a7d20ab13069", "doi_or_identifier": "10.52710/cfs.1008", "venue_or_site": "Computer fraud & security", "abstract_or_summary": "Semantic heterogeneity is one of the most difficult challenges in EAI․ Integration errors, reconciliation issues, and misalignment of business objectives can occur when integrating heterogeneous data sources with different data models, business terms, master data definitions, and process semantics, especially in the situation of complex hybrid SAP and multi-cloud landscapes․ We present a Generative AI (GenAI) based semantic integration architecture for SAP landscapes․ The solution uses LLMs, enterprise knowledge graphs, and SAP cloud integration services to enable context-aware, business-aligned interoperability at scale through generative AI-improved design, mapping, orchestration, and governance for SAP's cloud environments․ It shifts SAP's interoperability architecture from being merely about syntax-based data exchange to a semantic interoperability layer that understands and relates business meaning, intent, and process to a variety of distributed SAP and non-SAP systems․", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "commercial", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 44 }, { "id": "s2-10-1109-igarss52108-2023-10281958", "title": "Towards Geospatial Knowledge Graph Infused Neuro-Symbolic AI for Remote Sensing Scene Understanding", "authors_or_org": "Abhishek Potnis, D. Lunga, Alexandre Sorokine, P. Dias, Lexie Yang, Jacob Arndt, Jordan Bowman, Jason Wohlgemuth", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://www.semanticscholar.org/paper/bb3ba3433a9df4f891798d2e5397e315752e7df6", "doi_or_identifier": "10.1109/igarss52108.2023.10281958", "venue_or_site": "IEEE International Geoscience and Remote Sensing Symposium", "abstract_or_summary": "Deep learning has proven its effectiveness in numerous tasks for remote sensing scene understanding. However there is an increasing interest to explore fusion of domain-specific background information to the deep neural network to further improve its performance. Remote sensing researchers are also working towards developing models that generalize and adapt to multiple applications. Generalization challenges coupled with the scarcity of large corpora of high-quality noise-free labelled data, have together fueled an interest for leveraging background information. Knowledge graphs serve as excellent choice to represent domain-specific information in a structured, standardized and extensible manner. Integrating symbolic knowledge representations in the form of Knowledge Graph Embedding (KGE) to perform neuro-symbolic reasoning is an emerging research direction promising significant impacts. This vision paper seeks to position ideas and provoke early thoughts toward advancing neuro-symbolic artificial intelligence in the context of geospatial challenges. Specifically, it conceptualizes and elaborates on an architecture for infusing geospatial knowledge from knowledge graph in a deep neural network pipeline. As guiding case studies - land-use land-cover classification, object detection and instance segmentation can benefit from infusing spatio-contextual information with remote sensing imagery. The discussion further reflects on and articulates the challenges and explainable AI opportunities anticipated when scaling and maintaining large-scale geospatial knowledge graphs.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "ontology_ai", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 44 }, { "id": "oa-https-doi-org-10-1016-j-jii-2022-100383", "title": "Digital Twins: State of the art theory and practice, challenges, and open research questions", "authors_or_org": "Angira Sharma, Edward Elson Kosasih, Jie Zhang, Alexandra Brintrup, Anisoara Calinescu", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1016/j.jii.2022.100383", "doi_or_identifier": "10.1016/j.jii.2022.100383", "venue_or_site": "Journal of Industrial Information Integration", "abstract_or_summary": "Digital Twin was introduced over a decade ago, as an innovative all-encompassing tool, with perceived benefits including real-time monitoring, simulation, optimisation and accurate forecasting. However, the theoretical framework and practical implementations of digital twin (DT) are yet to fully achieve this vision at scale. Although an increasing number of successful implementations exist in research and industrial works, sufficient implementation details are not publicly available, making it difficult to fully assess their components and effectiveness, to draw comparisons, identify successful solutions, share lessons, and thus to jointly advance and benefit from the DT methodology. This work first presents a review of relevant DT research and industrial works, focusing on the key DT features, current approaches in different domains, and successful DT implementations, to infer the key DT components and properties, and to identify current limitations and reasons behind the delay in the widespread implementation and adoption of digital twin. This work identifies that the major reasons for this delay are: the fact the DT is still a fast evolving concept; the lack of a universal DT reference framework, e.g. DT standards are scarce and still evolving; problem- and domain-dependence; security concerns over shared data; lack of DT performance metrics; and reliance of digital twin on other fast-evolving technologies. Advancements in machine learning, Internet of Things (IoT) and big data have led to significant improvements in DT features such as real-time monitoring and accurate forecasting. Despite this progress and individual company-based efforts, certain research and implementation gaps exist in the field, which have so far prevented the widespread adoption of the DT concept and technology; these gaps are also discussed in this work. Based on reviews of past work and the identified gaps, this work then defines a conceptualisation of DT which includes its components and properties; these also validate the uniqueness of DT as a concept, when compared to similar concepts such as simulation, autonomous systems and optimisation. Real-life case studies are used to showcase the application of the conceptualisation. This work discusses the state-of-the-art in DT, addresses relevant and timely DT questions, and identifies novel research questions, thus contributing to a better understanding of the DT paradigm and advancing the theory and practice of DT and its allied technologies.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "commercial", "digital-twin", "enterprise-ai", "governance", "implementation-gaps", "industry", "openalex", "reference-framework" ], "triage_tier": "candidate", "triage_score": 43 }, { "id": "s2-10-1016-j-modpat-2025-100705", "title": "Future of Artificial Intelligence (AI) - Machine Learning (ML) Trends in Pathology and Medicine.", "authors_or_org": "Matthew G. Hanna, Liron Pantanowitz, Rajesh C Dash, J. Harrison, Mustafa Deebajah, J. Pantanowitz, H. Rashidi", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/92b21a25404ffeff52870cc13e590a1411404380", "doi_or_identifier": "10.1016/j.modpat.2025.100705", "venue_or_site": "Modern Pathology", "abstract_or_summary": "Artificial Intelligence (AI) and Machine Learning (ML) are transforming the field of medicine. Healthcare organizations are now starting to establish management strategies for integrating such platforms (AI-ML toolsets) which leverage the computational power of advanced algorithms to analyze data and to provide better insights which ultimately translates to enhanced clinical decision-making and improved patient outcomes. Emerging AI-ML platforms and trends in pathology and medicine are reshaping the field by offering innovative solutions to enhance diagnostic accuracy, operational workflows, clinical decision support, and clinical outcomes. These tools are also increasingly valuable in pathology research where they contribute to automated image analysis, biomarker discovery, drug development, clinical trials, and productive analytics. Other related trends include the adoption of ML-Ops (Machine Learning Operations) for managing models in clinical settings, the application of multimodal and multi-agent AI to utilize diverse data sources, expedited translational research and virtualized education for training and simulation. As the final chapter of our AI educational series, this review article delves into the current adoption, future directions, and transformative potential of AI-ML platforms in pathology and medicine, discussing their applications, benefits, challenges, and future perspectives.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 43 }, { "id": "ont-ai-030", "title": "A Semantic Loss Function for Deep Learning with Symbolic Knowledge", "authors_or_org": "Jingyi Xu; Zilu Zhang; Tal Friedman; Yitao Liang; Guy Van den Broeck", "year": 2018, "source_type": "paper", "bucket": "academic", "url": "https://proceedings.mlr.press/v80/xu18h.html", "doi_or_identifier": "pmlr v80", "venue_or_site": "ICML 2018", "abstract_or_summary": "Introduces semantic loss, a differentiable objective that penalizes neural predictions violating symbolic constraints.", "key_claims": [ "Symbolic constraints can be compiled into a differentiable learning objective.", "Semantic loss can improve neural model adherence to known constraints.", "Constraint-aware learning bridges logical knowledge and neural optimization." ], "ontology_relevance": "Shows how ontology-like constraints can shape model learning.", "ai_relevance": "Important method for integrating logic with deep learning.", "palantir_relevance": "Relevant to enforcing operational rules in ML models that act over ontology objects.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "deep-learning", "neuro-symbolic", "semantic-loss", "symbolic-constraints" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase6-xu-2025-a-mem-agentic-memory", "title": "A-Mem: Agentic Memory for LLM Agents", "authors_or_org": "Wujiang Xu; Zujie Liang; Kai Mei; Hang Gao; Juntao Tan; Yongfeng Zhang", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://openreview.net/forum?id=FiM0M8gcct", "doi_or_identifier": "openreview:fim0m8gcct; arxiv:2502.12110", "venue_or_site": "NeurIPS 2025 / OpenReview", "abstract_or_summary": "Proposes an agentic memory system that dynamically creates, links, and evolves structured notes for LLM agents.", "key_claims": [ "Memory should be actively organized and updated, not only retrieved.", "Zettelkasten-like linked memory improves long-term agent performance in the reported tasks.", "New memories can trigger link generation and evolution of old memories." ], "ontology_relevance": "Indirect but useful for ontology-like memory graphs, structured attributes, and evolving semantic networks.", "ai_relevance": "Strong source for LLM agent memory architecture and long-horizon task support.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "agent-memory", "llm-agents", "long-term-memory", "memory-graph", "neurips-2025", "phase6", "zettelkasten" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-std-nist-ai-100-3-2024", "title": "Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations", "authors_or_org": "Apostol Vassilev, Alina Oprea, Alie Fordyce, Hyrum Anderson", "year": 2024, "source_type": "technical_report", "bucket": "technical", "url": "https://doi.org/10.6028/NIST.AI.100-3", "doi_or_identifier": "10.6028/nist.ai.100-3", "venue_or_site": "NIST AI 100-3", "abstract_or_summary": "NIST taxonomy and terminology for adversarial machine learning attacks and mitigations, covering evasion, poisoning, privacy, abuse, and system-level threats.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "adversarial-ml", "ai-assurance", "controls", "nist", "risk-management", "security", "taxonomy" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-wand-weber-1990-ontological-model-is", "title": "An Ontological Model of an Information System", "authors_or_org": "Yair Wand and Ron Weber", "year": 1990, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1109/32.55073", "doi_or_identifier": "10.1109/32.55073", "venue_or_site": "IEEE Transactions on Software Engineering", "abstract_or_summary": "Classic information-systems paper using ontology to analyze what information systems represent and how representation quality can be evaluated.", "key_claims": [ "Information systems can be analyzed as representations of real-world systems.", "Ontological analysis can expose deficits such as construct overload, redundancy, and ambiguity.", "Conceptual modeling quality depends on whether system constructs map coherently to domain phenomena." ], "ontology_relevance": "Connects ontology to information-system design rather than only Semantic Web standards.", "ai_relevance": "AI agents acting through information systems inherit the representational commitments and defects of those systems.", "palantir_relevance": "Directly relevant to operational ontology as a model of enterprise objects, events, and permissible state changes.", "quality_signal": "widely_cited", "retrieval_tags": [ "conceptual-modeling", "enterprise-ontology", "information-systems", "phase3", "representation-quality" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase6-lippolis-2025-domain-specific-ontology-generation-llms", "title": "Assessing the Capability of Large Language Models for Domain-Specific Ontology Generation", "authors_or_org": "Anna Sofia Lippolis; Mohammad Javad Saeedizade; Robin Keskisarkka; Aldo Gangemi; Eva Blomqvist; Andrea Giovanni Nuzzolese", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2504.17402", "doi_or_identifier": "10.48550/arxiv.2504.17402", "venue_or_site": "ELMKE workshop at ESWC 2025 / arXiv", "abstract_or_summary": "Tests whether reasoning LLMs can generalize ontology generation across six domains using competency questions and user stories.", "key_claims": [ "Reasoning LLMs show promise for scalable domain-agnostic ontology construction in the evaluated settings.", "Performance is relatively consistent across tested domains.", "Evaluation uses curated competency questions across existing ontology engineering projects." ], "ontology_relevance": "Useful for domain ontology generation, competency-question-driven design, and cross-domain generalization.", "ai_relevance": "Compares reasoning-oriented LLMs for ontology modeling behavior across domains.", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "competency-questions", "domain-ontology", "elmke-2025", "llm", "ontology-generation", "phase6", "reasoning-llm" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-barocas-selbst-2016-disparate-impact", "title": "Big Data's Disparate Impact", "authors_or_org": "Solon Barocas and Andrew D. Selbst", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.15779/Z38BG31", "doi_or_identifier": "10.15779/z38bg31", "venue_or_site": "California Law Review", "abstract_or_summary": "Legal and technical analysis of how data mining can produce discriminatory outcomes even without explicit protected-class variables.", "key_claims": [ "Data-driven systems can reproduce inequality through target variables, training data, feature selection, and proxy variables.", "Discrimination can arise from apparently neutral data and classification practices.", "Legal accountability struggles when harms emerge from complex data pipelines." ], "ontology_relevance": "Shows that ontology categories and relations can encode proxies, targets, and historical inequalities.", "ai_relevance": "Essential critique for AI systems that classify people or allocate opportunities using integrated data.", "palantir_relevance": "Relevant where operational ontology supports consequential decisions in policing, health, welfare, or employment contexts.", "quality_signal": "widely_cited", "retrieval_tags": [ "ai-governance", "algorithmic-discrimination", "classification", "law", "phase3", "proxy-variables" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "lenat_guha_1989_large_kbs", "title": "Building Large Knowledge-Based Systems: Representation and Inference in the Cyc Project", "authors_or_org": "Douglas B. Lenat and R. V. Guha", "year": 1989, "source_type": "book", "bucket": "books", "url": "https://dl.acm.org/doi/book/10.5555/575523", "doi_or_identifier": "isbn 9780201517521; acm 10.5555/575523", "venue_or_site": "Addison-Wesley", "abstract_or_summary": "Book-length account of Cyc's approach to representing common-sense knowledge, organizing contexts, and supporting inference.", "key_claims": [ "Large AI knowledge bases need explicit representation choices and inference architecture.", "Context and assumptions must be represented, not left implicit.", "Knowledge engineering at scale requires modularization and maintenance discipline." ], "ontology_relevance": "Major source on large-scale formal knowledge-base construction.", "ai_relevance": "Provides lessons on the costs and benefits of explicit knowledge for AI.", "palantir_relevance": "Relevant to enterprise-scale ontology maintenance, contextual knowledge, and reasoning over operations.", "quality_signal": "widely_cited", "retrieval_tags": [ "context", "cyc", "inference", "knowledge-based-systems", "representation" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "arxiv-1104-4966v1", "title": "Combining Ontology Development Methodologies and Semantic Web Platforms for E-government Domain Ontology Development", "authors_or_org": "Jean Vincent Fonou Dombeu, Magda Huisman", "year": 2011, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/1104.4966v1", "doi_or_identifier": "1104.4966v1", "venue_or_site": "arXiv", "abstract_or_summary": "One of the key challenges in electronic government (e-government) is the development of systems that can be easily integrated and interoperated to provide seamless services delivery to citizens. In recent years, Semantic Web technologies based on ontology have emerged as promising solutions to the above engineering problems. However, current research practicing semantic development in e-government does not focus on the application of available methodologies and platforms for developing government domain ontologies. Furthermore, only a few of these researches provide detailed guidelines for developing semantic ontology models from a government service domain. This research presents a case study combining an ontology building methodology and two state-of-the-art Semantic Web platforms namely Protege and Java Jena ontology API for semantic ontology development in e-government. Firstly, a framework adopted from the Uschold and King ontology building methodology is employed to build a domain ontology describing the semantic content of a government service domain. Thereafter, UML is used to semi-formally represent the domain ontology. Finally, Protege and Jena API are employed to create the Web Ontology Language (OWL) and Resource Description Framework (RDF) representations of the domain ontology respectively to enable its computer processing. The study aims at: (1) providing e-government developers, particularly those from the developing world with detailed guidelines for practicing semantic content development in their e-government projects and (2), strengthening the adoption of semantic technologies in e-government. The study would also be of interest to novice Semantic Web developers who might used it as a starting point for further investigations.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "foundational" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "ont-ai-054", "title": "ConceptNet 5.5: An Open Multilingual Graph of General Knowledge", "authors_or_org": "Robyn Speer; Joshua Chin; Catherine Havasi", "year": 2017, "source_type": "paper", "bucket": "academic", "url": "https://ojs.aaai.org/index.php/AAAI/article/view/11164", "doi_or_identifier": "aaai 2017", "venue_or_site": "AAAI 2017", "abstract_or_summary": "Describes ConceptNet 5.5, a multilingual commonsense knowledge graph assembled from multiple resources.", "key_claims": [ "Commonsense knowledge can be represented as a multilingual graph of concepts and relations.", "Combining multiple knowledge sources improves coverage.", "Commonsense KGs support NLP and reasoning tasks but include noise and cultural assumptions." ], "ontology_relevance": "Important commonsense graph with ontology-like relations.", "ai_relevance": "Used for commonsense QA, embeddings, and knowledge-augmented NLP.", "palantir_relevance": "Indirectly relevant as background knowledge; domain-specific enterprise concepts still require local ontologies.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "commonsense", "conceptnet", "knowledge-graph", "multilingual" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-palantir-aip-bootcamps-2023", "title": "Deploying Full Spectrum AI in Days: How AIP Bootcamps Work", "authors_or_org": "Palantir Technologies", "year": 2023, "source_type": "blog", "bucket": "palantir", "url": "https://blog.palantir.com/deploying-full-spectrum-ai-in-days-how-aip-bootcamps-work-21829ec8d560", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir blog explaining AIP Bootcamps, a go-to-market and implementation method for rapidly building operational AI workflows on customer data and ontology structures.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "marketing", "retrieval_tags": [ "aip", "bootcamp", "commercial-claim", "implementation-method", "palantir" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "p2-llm-ont-003", "title": "Do LLMs Really Adapt to Domains? An Ontology Learning Perspective", "authors_or_org": "Huu Tan Mai; Cuong Xuan Chu; Heiko Paulheim", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2407.19998", "doi_or_identifier": "arxiv:2407.19998", "venue_or_site": "ISWC 2024 / arXiv", "abstract_or_summary": "Uses controlled ontology learning experiments to test whether LLMs reason over domain relationships or rely on lexical patterns when domain terms are arbitrary or unfamiliar.", "key_claims": [ "Off-the-shelf LLMs can mistake familiar lexical senses for domain reasoning.", "Fine-tuning improves ontology learning behavior when terms are arbitrary or domain-specific.", "Domain adaptation claims should be tested with controlled semantic perturbations." ], "ontology_relevance": "Important caution for using LLMs to learn domain ontologies from unfamiliar enterprise vocabulary.", "ai_relevance": "Distinguishes lexical pattern matching from robust semantic adaptation in LLM behavior.", "palantir_relevance": "Relevant to enterprise ontologies where local object labels may not match public web meanings.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "domain-adaptation", "iswc-2024", "llm", "ontology-learning", "taxonomy" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "ont-ai-016", "title": "ERNIE: Enhanced Language Representation with Informative Entities", "authors_or_org": "Zhengyan Zhang; Xu Han; Zhiyuan Liu; Xin Jiang; Maosong Sun; Qun Liu", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://aclanthology.org/P19-1139", "doi_or_identifier": "10.18653/v1/p19-1139", "venue_or_site": "ACL 2019", "abstract_or_summary": "Introduces an entity-enhanced language representation model that integrates textual and knowledge graph information.", "key_claims": [ "Entity information from KGs can improve language representation.", "Jointly modeling text and entities helps knowledge-driven NLP tasks.", "Knowledge integration requires entity alignment between text and graph." ], "ontology_relevance": "Shows how graph entities can ground language models in named concepts.", "ai_relevance": "Early influential knowledge-enhanced pretraining approach.", "palantir_relevance": "Relevant to tying language interfaces to enterprise ontology objects.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "entity-linking", "knowledge-enhanced-lm", "knowledge-graph", "pretraining" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "arxiv-2404-17000v1", "title": "Evaluating Class Membership Relations in Knowledge Graphs using Large Language Models", "authors_or_org": "Bradley P. Allen, Paul T. Groth", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2404.17000v1", "doi_or_identifier": "2404.17000v1", "venue_or_site": "arXiv", "abstract_or_summary": "A backbone of knowledge graphs are their class membership relations, which assign entities to a given class. As part of the knowledge engineering process, we propose a new method for evaluating the quality of these relations by processing descriptions of a given entity and class using a zero-shot chain-of-thought classifier that uses a natural language intensional definition of a class. We evaluate the method using two publicly available knowledge graphs, Wikidata and CaLiGraph, and 7 large language models. Using the gpt-4-0125-preview large language model, the method's classification performance achieves a macro-averaged F1-score of 0.830 on data from Wikidata and 0.893 on data from CaLiGraph. Moreover, a manual analysis of the classification errors shows that 40.9% of errors were due to the knowledge graphs, with 16.0% due to missing relations and 24.9% due to incorrectly asserted relations. These results show how large language models can assist knowledge engineers in the process of knowledge graph refinement. The code and data are available on Github.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-nature-patil-2023-gorilla", "title": "Gorilla: Large Language Model Connected with Massive APIs", "authors_or_org": "Shishir G. Patil, Tianjun Zhang, Xin Wang, Joseph E. Gonzalez", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2305.15334", "doi_or_identifier": "10.48550/arxiv.2305.15334", "venue_or_site": "arXiv", "abstract_or_summary": "Studies LLM API calling over large tool collections, highlighting tool retrieval, grounding, and hallucination risks in tool-using systems.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "api", "hallucination", "llm-agents", "retrieval", "tool-use" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "arxiv-2508-05498v2", "title": "GRAIL:Learning to Interact with Large Knowledge Graphs for Retrieval Augmented Reasoning", "authors_or_org": "Ge Chang, Jinbo Su, Jiacheng Liu, Pengfei Yang, Yuhao Shang, Huiwen Zheng, Hongli Ma, Yan Liang", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/2508.05498v2", "doi_or_identifier": "2508.05498v2", "venue_or_site": "arXiv", "abstract_or_summary": "Large Language Models (LLMs) integrated with Retrieval-Augmented Generation (RAG) techniques have exhibited remarkable performance across a wide range of domains. However, existing RAG approaches primarily operate on unstructured data and demonstrate limited capability in handling structured knowledge such as knowledge graphs. Meanwhile, current graph retrieval methods fundamentally struggle to capture holistic graph structures while simultaneously facing precision control challenges that manifest as either critical information gaps or excessive redundant connections, collectively undermining reasoning performance. To address this challenge, we propose GRAIL: Graph-Retrieval Augmented Interactive Learning, a framework designed to interact with large-scale graphs for retrieval-augmented reasoning. Specifically, GRAIL integrates LLM-guided random exploration with path filtering to establish a data synthesis pipeline, where a fine-grained reasoning trajectory is automatically generated for each task. Based on the synthesized data, we then employ a two-stage training process to learn a policy that dynamically decides the optimal actions at each reasoning step. The overall objective of precision-conciseness balance in graph retrieval is decoupled into fine-grained process-supervised rewards to enhance data efficiency and training stability. In practical deployment, GRAIL adopts an interactive retrieval paradigm, enabling the model to autonomously explore graph paths while dynamically balancing retrieval breadth and precision. Extensive experiments have shown that GRAIL achieves an average accuracy improvement of 21.01% and F1 improvement of 22.43% on three knowledge graph question-answering datasets. Our source code and datasets is available at https://github.com/Changgeww/GRAIL.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "ontology_ai" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "arxiv-1601-02650v1", "title": "Inference rules for RDF(S) and OWL in N3Logic", "authors_or_org": "Dominik Tomaszuk", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/1601.02650v1", "doi_or_identifier": "1601.02650v1", "venue_or_site": "arXiv", "abstract_or_summary": "This paper presents inference rules for Resource Description Framework (RDF), RDF Schema (RDFS) and Web Ontology Language (OWL). Our formalization is based on Notation 3 Logic, which extended RDF by logical symbols and created Semantic Web logic for deductive RDF graph stores. We also propose OWL-P that is a lightweight formalism of OWL and supports soft inferences by omitting complex language constructs.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "foundational" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-palantir-aipcon8-demos-2025", "title": "Inside the AIPCon 8 Demos Transforming Manufacturing, Insurance, and Construction", "authors_or_org": "Palantir Technologies", "year": 2025, "source_type": "blog", "bucket": "palantir", "url": "https://blog.palantir.com/inside-the-aipcon-8-demos-transforming-manufacturing-insurance-and-construction-2ef01d53ea96", "doi_or_identifier": null, "venue_or_site": "Palantir Blog", "abstract_or_summary": "Palantir blog summarizing AIPCon demonstrations across sectors, useful for identifying how Palantir publicly presents ontology-grounded AI workflows to customers.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "marketing", "retrieval_tags": [ "aip", "aipcon", "customer-demos", "ontology", "palantir" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "ont-ai-045", "title": "LUBM: A Benchmark for OWL Knowledge Base Systems", "authors_or_org": "Yuanbo Guo; Zhengxiang Pan; Jeff Heflin", "year": 2005, "source_type": "paper", "bucket": "academic", "url": "https://www.sciencedirect.com/science/article/pii/S1570826805000132", "doi_or_identifier": "10.1016/j.websem.2005.06.005", "venue_or_site": "Journal of Web Semantics", "abstract_or_summary": "Introduces the Lehigh University Benchmark for evaluating OWL knowledge base systems using a university domain ontology and synthetic data.", "key_claims": [ "OWL systems need benchmarks that test reasoning and query performance.", "Synthetic data generation enables scaling evaluations.", "Benchmark design must reflect ontology expressivity and query workload." ], "ontology_relevance": "Classic benchmark for ontology reasoning and storage systems.", "ai_relevance": "Relevant to evaluating knowledge infrastructure used by AI systems.", "palantir_relevance": "Useful analogy for evaluating enterprise ontology scale and query performance.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "benchmark", "evaluation", "lubm", "owl", "reasoning" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-llm-ontmem-013", "title": "Magneto: Combining Small and Large Language Models for Schema Matching", "authors_or_org": "Yurong Liu; Eduardo Pena; Aecio S. R. Santos; Eden Wu; Juliana Freire", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.14778/3742728.3742757", "doi_or_identifier": "10.14778/3742728.3742757", "venue_or_site": "Proceedings of the VLDB Endowment", "abstract_or_summary": "Proposes Magneto, combining small and large language models for schema matching to balance cost, accuracy, and deployment practicality.", "key_claims": [ "Schema matching can combine cheaper small models with selective large-model use instead of relying on expensive LLM calls for every pair.", "Cost-aware model orchestration matters for large-scale semantic alignment.", "Benchmarks with mostly lexical matches can overstate performance and should be interpreted carefully." ], "ontology_relevance": "Relevant to scalable ontology and semantic-layer alignment where exhaustive LLM comparison is impractical.", "ai_relevance": "Supports hybrid small/large model orchestration for structured matching workflows.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "cost-aware", "large-language-model", "schema-matching", "small-language-model", "vldb" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase6-xu-2025-memq-kg-reasoning", "title": "Memory-Augmented Query Reconstruction for LLM-Based Knowledge Graph Reasoning", "authors_or_org": "Mufan Xu; Gewen Liang; Kehai Chen; Wei Wang; Xun Zhou; Muyun Yang; Tiejun Zhao; Min Zhang", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://aclanthology.org/2025.findings-acl.1234", "doi_or_identifier": "10.18653/v1/2025.findings-acl.1234", "venue_or_site": "Findings of ACL 2025", "abstract_or_summary": "Introduces MemQ, which stores explicit query memories to reconstruct knowledge-graph queries and separate LLM reasoning from tool invocation.", "key_claims": [ "Memory of query statements improves KG reasoning workflows.", "Decoupling reasoning from tool invocation can make KGQA more reliable.", "Readable reasoning steps help LLM-based KG reasoning." ], "ontology_relevance": "Relevant to ontology/KG query patterns, semantic parsing, and query-memory reuse.", "ai_relevance": "Connects LLM reasoning, memory modules, and KG query reconstruction.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "acl-2025", "kgqa", "knowledge-graph", "llm-memory", "phase6", "query-reconstruction", "semantic-parsing" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-mitchell-etal-2019-model-cards", "title": "Model Cards for Model Reporting", "authors_or_org": "Margaret Mitchell, Simone Wu, Andrew Zaldivar, Parker Barnes, Lucy Vasserman, Ben Hutchinson, Elena Spitzer, Inioluwa Deborah Raji, and Timnit Gebru", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3287560.3287596", "doi_or_identifier": "10.1145/3287560.3287596", "venue_or_site": "ACM Conference on Fairness, Accountability, and Transparency", "abstract_or_summary": "Introduces model cards as structured documentation for model details, intended use, factors, metrics, evaluation data, and ethical considerations.", "key_claims": [ "Structured model documentation improves transparency and responsible use.", "Evaluation should be reported across intended uses, factors, metrics, and populations.", "Documentation supports accountability but must be kept current and connected to deployment context." ], "ontology_relevance": "Suggests analogous documentation needs for ontology modules, object types, action types, mappings, and constraints.", "ai_relevance": "Core AI documentation source for model governance and deployment transparency.", "palantir_relevance": "Useful comparator for documenting AIP models and ontology-exposed tools.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ai-governance", "documentation", "evaluation", "model-cards", "phase3", "transparency" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-llm-ontmem-002", "title": "NeOn-GPT: A Large Language Model-Powered Pipeline for Ontology Learning", "authors_or_org": "Nadeen Shoukry Fathallah; Arunav Das; Stefano De Giorgis; Andrea Poltronieri; Peter Haase; Liubov Kovriguina", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/abs/10.1007/978-3-031-78952-6_4", "doi_or_identifier": "10.1007/978-3-031-78952-6_4", "venue_or_site": "Extended Semantic Web Conference 2024", "abstract_or_summary": "Presents NeOn-GPT, a pipeline combining the NeOn ontology engineering methodology with LLM generation, implemented on metaphactory and evaluated with the Stanford wine ontology as a gold standard.", "key_claims": [ "LLM-based ontology learning benefits from being embedded in an explicit ontology engineering methodology.", "A workflow pipeline is more controllable than asking a model to generate an ontology in a single step.", "Evaluation against a gold ontology exposes both useful automation and remaining limits in reasoning, fact checking, and domain expertise." ], "ontology_relevance": "Concrete example of wrapping LLM generation in a recognized ontology engineering methodology.", "ai_relevance": "Shows how LLM prompting can be operationalized as part of a structured ontology learning pipeline.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "eswc-2024", "llm", "neon-methodology", "ontology-learning", "pipeline" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-palantir-digitalhealth-fdp-award-2023", "title": "NHS England awards 330m Federated Data Platform contract to Palantir", "authors_or_org": "Digital Health", "year": 2023, "source_type": "news", "bucket": "palantir", "url": "https://www.digitalhealth.net/2023/11/nhs-england-awards-330m-federated-data-platform-contract-to-palantir", "doi_or_identifier": null, "venue_or_site": "Digital Health", "abstract_or_summary": "Trade-press report on NHS England awarding the Federated Data Platform contract to Palantir, useful for chronology and contract-value context.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "secondary_source", "retrieval_tags": [ "contract-award", "fdp", "independent-reporting", "nhs", "palantir", "procurement" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-std-nist-privacy-framework-1-0-2020", "title": "NIST Privacy Framework: A Tool for Improving Privacy through Enterprise Risk Management, Version 1.0", "authors_or_org": "National Institute of Standards and Technology", "year": 2020, "source_type": "framework", "bucket": "technical", "url": "https://www.nist.gov/privacy-framework/privacy-framework", "doi_or_identifier": null, "venue_or_site": "NIST Privacy Framework", "abstract_or_summary": "NIST framework for managing privacy risk through organizational profiles, core functions, categories, and subcategories aligned with enterprise risk management.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "data-governance", "enterprise-controls", "nist", "policy", "privacy", "risk-management" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-ols4-ontology-lookup-service-2025", "title": "OLS4: a new Ontology Lookup Service for a growing interdisciplinary knowledge ecosystem", "authors_or_org": "OLS4 contributors / EMBL-EBI", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/bioinformatics/btaf279", "doi_or_identifier": "10.1093/bioinformatics/btaf279", "venue_or_site": "Bioinformatics", "abstract_or_summary": "Introduces OLS4, a redesigned Ontology Lookup Service for searching, browsing, and programmatically accessing a growing interdisciplinary ecosystem of ontologies.", "key_claims": [ "Ontology services need scalable search, metadata, term lookup, and API access.", "A growing interdisciplinary ontology ecosystem requires infrastructure beyond individual ontology files.", "Programmatic ontology access supports annotation, integration, and AI pipelines." ], "ontology_relevance": "Direct evidence that ontology ecosystems require service layers for discovery and machine access.", "ai_relevance": "Ontology lookup services are core infrastructure for entity normalization, semantic search, annotation, and agent access to controlled vocabularies.", "palantir_relevance": "Useful comparator for ontology catalog, discovery, and agent tool exposure in enterprise platforms.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "api", "biomedical-ontology", "ols4", "ontology-lookup-service", "ontology-repository", "phase7", "services" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase2-ontogenix-2025", "title": "OntoGenix: Leveraging Large Language Models for enhanced ontology engineering from datasets", "authors_or_org": "Mikel Val-Calvo; Mikel Egana Aranguren; Juan Mulero-Hernandez; Gines Almagro-Hernandez; Jesualdo Tomas Fernandez-Breis", "year": 2025, "source_type": "paper", "bucket": "academic", "url": "https://dl.acm.org/doi/10.1016/j.ipm.2024.104042", "doi_or_identifier": "10.1016/j.ipm.2024.104042", "venue_or_site": "Information Processing & Management", "abstract_or_summary": "Peer-reviewed article on using LLMs to enhance ontology engineering from datasets, positioned within a special issue on ontologies and LLMs.", "key_claims": [ "Datasets can be treated as inputs for LLM-supported ontology engineering rather than only as data to annotate.", "LLM assistance needs to be evaluated as part of ontology engineering methodology, not only NLP performance.", "Ontology generation from datasets connects data profiling, conceptual modeling, and semantic enrichment." ], "ontology_relevance": "Peer-reviewed evidence for dataset-driven LLM ontology engineering.", "ai_relevance": "Useful for practical pipelines that turn existing enterprise/scientific data into ontology structures.", "palantir_relevance": "Comparable to ontology extraction from integrated enterprise datasets.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "dataset", "llm", "ontology-engineering", "peer-reviewed", "semantic-enrichment" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-ro-crate-packaging-research-artefacts-2022", "title": "Packaging research artefacts with RO-Crate", "authors_or_org": "Stian Soiland-Reyes; Peter Sefton; Mercè Crosas; Leyla Jael Castro; et al.", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.3233/ds-210053", "doi_or_identifier": "10.3233/ds-210053", "venue_or_site": "Data Science", "abstract_or_summary": "RO-Crate packages research objects and metadata using lightweight linked-data conventions, making datasets, software, workflows, people, and provenance more portable and machine-readable.", "key_claims": [ "Research artifacts need portable machine-readable metadata packages.", "RO-Crate uses linked-data conventions to describe datasets, software, workflows, people, and provenance.", "Packaging metadata with artifacts improves reuse and automated processing." ], "ontology_relevance": "Connects semantic web, provenance, and FAIR data principles to practical research artifact packaging.", "ai_relevance": "Gives AI agents a practical model for consuming scientific artifacts with metadata, provenance, and executable workflow context.", "palantir_relevance": "Relevant to data-product packaging, provenance, and machine-readable operational context in enterprise AI platforms.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "fair-data", "linked-data", "metadata", "phase7", "provenance", "research-object", "ro-crate", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "pal-impact-tampa-general-2026", "title": "Palantir Impact: Tampa General Hospital", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "case_study", "bucket": "palantir", "url": "https://www.palantir.com/impact/tampa-general-hospital", "doi_or_identifier": null, "venue_or_site": "Palantir Impact", "abstract_or_summary": "Vendor impact page describing Tampa General Hospital's use of Foundry Ontology to integrate hospital data sources such as schedules and census data for operational workflows.", "key_claims": [ "Tampa General integrates nurse schedules, patient census, surgery schedules, and other sources into Foundry Ontology.", "Palantir presents the deployment as supporting hospital operations and decision-making.", "The case is framed as a healthcare operational AI/data platform example." ], "ontology_relevance": "Applied example of modeling hospital operations in Foundry Ontology.", "ai_relevance": "Shows how integrated operational context can support AI-enabled healthcare workflows.", "palantir_relevance": "Vendor-provided customer evidence; useful but not independently audited in this source.", "quality_signal": "marketing", "retrieval_tags": [ "case-study", "foundry-ontology", "healthcare", "hospital", "palantir", "tampa-general" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-mittelstadt-2019-principles-alone", "title": "Principles Alone Cannot Guarantee Ethical AI", "authors_or_org": "Brent Mittelstadt", "year": 2019, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/s42256-019-0114-4", "doi_or_identifier": "10.1038/s42256-019-0114-4", "venue_or_site": "Nature Machine Intelligence", "abstract_or_summary": "Commentary arguing that AI ethics principles are insufficient without professional norms, accountability mechanisms, and institutional implementation.", "key_claims": [ "Ethical principles are often too abstract to guide concrete AI practice.", "AI needs accountability institutions, professional responsibilities, and enforcement mechanisms.", "Principles can create a false sense of governance if not operationalized." ], "ontology_relevance": "Important warning that ontology governance principles must become enforceable workflows, permissions, review, and audit.", "ai_relevance": "Directly relevant to article framing around moving from principles to operational AI controls.", "palantir_relevance": "Useful for evaluating whether platform governance claims are operational controls or high-level assurances.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "accountability", "ai-ethics", "governance", "nature", "phase3", "principles-to-practice" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase2-ragas-2024", "title": "RAGAs: Automated Evaluation of Retrieval Augmented Generation", "authors_or_org": "Shahul Es; Jithin James; Luis Espinosa-Anke; Steven Schockaert", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2309.15217", "doi_or_identifier": "arxiv:2309.15217", "venue_or_site": "EACL 2024 System Demonstrations", "abstract_or_summary": "Introduces RAGAS, a reference-free evaluation framework for RAG pipelines with metrics for retrieval and generation quality.", "key_claims": [ "RAG systems need metrics that separately assess context relevance, faithfulness, and answer relevance.", "Reference-free evaluation can support iterative RAG development when gold answers are scarce.", "Ontology-grounded RAG should be evaluated at retrieval, grounding, and answer levels.", "RAG evaluation should measure both retrieval context and generated answer quality.", "Faithfulness can be evaluated against retrieved context rather than only against labels.", "Automated RAG metrics are useful but should be calibrated for domain-specific stakes." ], "ontology_relevance": "Supports evaluation design for ontology-backed retrieval systems.", "ai_relevance": "Important RAG evaluation tool and paper for production-quality LLM systems.", "palantir_relevance": "Comparable to AIP Evals, but oriented toward general RAG pipeline assessment.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "answer-correctness", "benchmark", "context-relevance", "evaluation", "faithfulness", "rag", "rag-evaluation", "ragas", "retrieval", "trustworthy-ai" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase2-graphrag-021", "title": "Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection", "authors_or_org": "Akari Asai; Zeqiu Wu; Yizhong Wang; Avirup Sil; Hannaneh Hajishirzi", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2310.11511", "doi_or_identifier": "arxiv:2310.11511", "venue_or_site": "ICLR 2024 / arXiv", "abstract_or_summary": "Introduces Self-RAG, a framework where a model learns when to retrieve, how to use retrieved passages, and how to critique generated output via reflection tokens.", "key_claims": [ "RAG systems should decide when retrieval is needed rather than always retrieving.", "Generated answers need critique for support and utility, not only fluent generation.", "Retrieval-aware control can reduce unsupported answers." ], "ontology_relevance": "Ontology-grounded systems can use schema and action constraints as stronger retrieval and critique signals.", "ai_relevance": "Key source for adaptive retrieval and self-critique in RAG systems.", "palantir_relevance": "Relevant to enterprise agent controls that decide when evidence or governed actions are required.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "adaptive-retrieval", "control", "critique", "faithfulness", "self-rag" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "arxiv-1111-1941v1", "title": "Semantic-Driven e-Government: Application of Uschold and King Ontology Building Methodology for Semantic Ontology Models Development", "authors_or_org": "Jean Vincent Fonou-Dombeu, Magda Huisman", "year": 2011, "source_type": "paper", "bucket": "academic", "url": "http://arxiv.org/abs/1111.1941v1", "doi_or_identifier": "1111.1941v1", "venue_or_site": "arXiv", "abstract_or_summary": "Electronic government (e-government) has been one of the most active areas of ontology development during the past six years. In e-government, ontologies are being used to describe and specify e-government services (e-services) because they enable easy composition, matching, mapping and merging of various e-government services. More importantly, they also facilitate the semantic integration and interoperability of e-government services. However, it is still unclear in the current literature how an existing ontology building methodology can be applied to develop semantic ontology models in a government service domain. In this paper the Uschold and King ontology building methodology is applied to develop semantic ontology models in a government service domain. Firstly, the Uschold and King methodology is presented, discussed and applied to build a government domain ontology. Secondly, the domain ontology is evaluated for semantic consistency using its semi-formal representation in Description Logic. Thirdly, an alignment of the domain ontology with the Descriptive Ontology for Linguistic and Cognitive Engineering (DOLCE) upper level ontology is drawn to allow its wider visibility and facilitate its integration with existing metadata standard. Finally, the domain ontology is formally written in Web Ontology Language (OWL) to enable its automatic processing by computers. The study aims to provide direction for the application of existing ontology building methodologies in the Semantic Web development processes of e-government domain specific ontology models; which would enable their repeatability in other e-government projects and strengthen the adoption of semantic technologies in e-government.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "preprint", "retrieval_tags": [ "arxiv", "foundational" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-llm-ontmem-003", "title": "Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES): A Method for Populating Knowledge Bases Using Zero-Shot Learning", "authors_or_org": "J. Harry Caufield; Harshad Hegde; Vincent Emonet; Nomi L. Harris; Marcin P. Joachimiak; Nicolas Matentzoglu; HyeongSik Kim; Sierra A. T. Moxon; Justin T. Reese; Melissa A. Haendel; Peter N. Robinson; Christopher J. Mungall", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://academic.oup.com/bioinformatics/article/40/3/btae104/7612230", "doi_or_identifier": "10.1093/bioinformatics/btae104", "venue_or_site": "Bioinformatics", "abstract_or_summary": "Introduces SPIRES, implemented in OntoGPT, for extracting structured knowledge from text into user-defined schemas using zero-shot LLM prompting and ontology-based grounding of identifiers.", "key_claims": [ "Schema-guided LLM extraction can populate complex nested knowledge models without task-specific training data.", "Grounding extracted elements to public ontology identifiers reduces ambiguity and supports curation.", "LLM extraction is useful as candidate knowledge acquisition, not a replacement for validation against curated resources." ], "ontology_relevance": "Important peer-reviewed source for ontology-grounded extraction into structured schemas and KGs.", "ai_relevance": "Demonstrates zero-shot LLM information extraction constrained by explicit schema and identifier normalization.", "palantir_relevance": "", "quality_signal": "peer_reviewed", "retrieval_tags": [ "knowledge-base-population", "ontogpt", "ontology-grounded-extraction", "spires", "zero-shot" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "ont-ai-035", "title": "The Debate over Understanding in AI's Large Language Models", "authors_or_org": "Melanie Mitchell; David C. Krakauer", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://www.pnas.org/doi/10.1073/pnas.2215907120", "doi_or_identifier": "10.1073/pnas.2215907120", "venue_or_site": "PNAS", "abstract_or_summary": "Reviews arguments about whether LLMs understand language and what kinds of tests and theories are needed to assess understanding.", "key_claims": [ "LLM behavior raises unresolved questions about understanding, abstraction, and grounding.", "Current evaluations often fail to settle deep questions about meaning and reasoning.", "Progress requires clearer definitions and more robust tests." ], "ontology_relevance": "Frames ontology as one possible mechanism for explicit reference and structure in AI systems.", "ai_relevance": "Current high-level synthesis of LLM understanding debates.", "palantir_relevance": "Useful for cautious framing of enterprise AI claims about understanding and reasoning.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "ai-debate", "evaluation", "grounding", "llm-understanding" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-human-phenotype-ontology-2021", "title": "The Human Phenotype Ontology in 2021", "authors_or_org": "Sebastian Kohler; Michael Gargano; Nicolas Matentzoglu; et al.", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/nar/gkaa1043", "doi_or_identifier": "10.1093/nar/gkaa1043", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "The 2021 HPO update summarizes ontology expansion, phenotype annotation, computational disease diagnosis, rare disease use, and integration with genomic resources.", "key_claims": [ "HPO structures clinical abnormalities for computation across diseases and genes.", "Phenotype annotations support diagnosis and translational research.", "Sustained curation and tooling are central to ontology utility." ], "ontology_relevance": "Canonical biomedical ontology source connecting formal terms to clinical data integration and computational reasoning.", "ai_relevance": "Supports the link between ontology, machine-readable clinical phenotype data, and AI-assisted diagnosis or variant prioritization.", "palantir_relevance": "Relevant as a non-commercial health-domain ontology precedent for consistent object modeling over complex medical data.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "clinical-ai", "diagnosis", "hpo", "ontology-infrastructure", "phase7", "phenotype" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-human-phenotype-ontology-2024", "title": "The Human Phenotype Ontology in 2024: phenotypes around the world", "authors_or_org": "Human Phenotype Ontology Consortium", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/nar/gkad1005", "doi_or_identifier": "10.1093/nar/gkad1005", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "Reports the 2024 state of the Human Phenotype Ontology, emphasizing globalized phenotypic descriptions, disease diagnostics, cross-language and cross-resource integration, and computational phenotype analysis.", "key_claims": [ "HPO provides structured phenotype terms for clinical and translational research.", "Phenotype ontologies enable cross-resource computation over patients, diseases, genes, and variants.", "Global adoption requires translations, mappings, tooling, and community curation." ], "ontology_relevance": "Strong domain ontology example linking human observations to computational diagnosis and research integration.", "ai_relevance": "Phenotype ontologies are widely used for diagnosis, variant prioritization, machine learning, and cross-database biomedical reasoning.", "palantir_relevance": "Analogous to operational object models where clinical features, patients, diseases, and evidence must be represented consistently.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "clinical-ai", "diagnosis", "hpo", "ontology-infrastructure", "phase7", "phenotype", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-monarch-initiative-2024", "title": "The Monarch Initiative in 2024: an analytic platform integrating phenotypes, genes and diseases across species", "authors_or_org": "The Monarch Initiative contributors", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1093/nar/gkad1082", "doi_or_identifier": "10.1093/nar/gkad1082", "venue_or_site": "Nucleic Acids Research", "abstract_or_summary": "Describes Monarch as an analytic platform integrating phenotypes, genes, genotypes, variants, diseases, and model organism knowledge across species.", "key_claims": [ "Ontology mappings enable integrated analysis of phenotypes, genes, diseases, and model organism data.", "Cross-species phenotype and disease integration supports translational discovery.", "Graph-based biomedical platforms require shared identifiers, ontologies, and curation workflows." ], "ontology_relevance": "High-value example of an ontology-backed scientific knowledge graph used for analysis rather than passive documentation.", "ai_relevance": "Demonstrates ontology-based integration for translational AI, cross-species inference, and graph-driven biomedical discovery.", "palantir_relevance": "Provides a scientific analogue for enterprise ontology platforms integrating heterogeneous operational entities.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "disease", "genes", "knowledge-graph", "monarch", "phase7", "phenotype", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase3-n3c-jamia-design-2021", "title": "The National COVID Cohort Collaborative (N3C): Rationale, design, infrastructure, and deployment", "authors_or_org": "Haendel et al.", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://academic.oup.com/jamia/article/28/3/427/5893482", "doi_or_identifier": "10.1093/jamia/ocaa196", "venue_or_site": "Journal of the American Medical Informatics Association", "abstract_or_summary": "Peer-reviewed paper describing the N3C data enclave, data harmonization, governance workstream, Data Use Request framework, access tiers, and deployment infrastructure, including Palantir Foundry as part of the hosted enclave environment.", "key_claims": [ "N3C developed a Data Use Request framework to protect patient data and support transparent access.", "The enclave uses access tiers including synthetic, de-identified, and limited datasets.", "The paper identifies Palantir Foundry as hosted within the N3C Enclave." ], "ontology_relevance": "Strong comparator for governed large-scale data integration and semantic harmonization.", "ai_relevance": "Provides a peer-reviewed health-data infrastructure model relevant to AI research governance.", "palantir_relevance": "Directly notes Palantir Foundry in a public-health data enclave, with governance context.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "data-enclave", "governance", "health-informatics", "jamia", "n3c", "palantir-foundry" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "n3c-jcts-design-infrastructure-2021", "title": "The National COVID Cohort Collaborative: Rationale, design, infrastructure, and deployment", "authors_or_org": "N3C Consortium authors", "year": 2021, "source_type": "paper", "bucket": "academic", "url": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7309060", "doi_or_identifier": null, "venue_or_site": "Journal of Clinical and Translational Science / PubMed Central", "abstract_or_summary": "Peer-reviewed description of the N3C data enclave rationale, design, infrastructure, deployment, harmonization, governance, and collaborative research environment.", "key_claims": [ "N3C built a shared clinical data enclave for COVID research across multiple contributing institutions.", "The paper describes governance, data harmonization, and secure analytic access as core infrastructure concerns.", "It is useful public evidence for how large health-data platforms handle access and research governance." ], "ontology_relevance": "Shows a healthcare case where integrated data, shared semantics, and governed analytic workspaces are operational requirements.", "ai_relevance": "Relevant to AI and ML research because N3C supports large-scale analytics under access controls.", "palantir_relevance": "N3C is a public-sector data enclave commonly associated with Palantir Foundry in public materials; use this paper for N3C governance and infrastructure, not vendor claims.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "clinical-data", "covid", "data-enclave", "governance", "infrastructure", "n3c" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "oa-https-doi-org-10-1080-01972243-2022-2100851", "title": "The seer and the seen: Surveying Palantir’s surveillance platform", "authors_or_org": "Andrew Iliadis, Amelia Acker", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://www.tandfonline.com/doi/full/10.1080/01972243.2022.2100851", "doi_or_identifier": "10.1080/01972243.2022.2100851", "venue_or_site": "The Information Society", "abstract_or_summary": "Palantir is among the most secretive and understudied surveillance firms globally. The company supplies information technology solutions for data integration and tracking to police and government agencies, humanitarian organizations, and corporations. To illuminate and learn more about Palantir’s opaque surveillance practices, we begin by sketching Palantir’s company history and contract network, followed by an explanation of key terms associated with Palantir’s technology area and a description of the firm’s platform ecosystem. We then summarize current scholarship on Palantir’s continuing role in policing, intelligence, and security operations. Our primary contribution and analysis are a computational topic modeling of Palantir’s surveillance patents (n = 155), including their topics and themes. We end by discussing the concept of infrastructuring to understand Palantir as a surveillance platform, where we theorize information standards like administrative metadata as phenomena for structuring social worlds in and through access to digital information.", "key_claims": [ "Palantir is analyzed through the lens of surveillance infrastructure.", "The article discusses government/policing clients and racial justice concerns around predictive policing and analytics.", "It frames Palantir as consequential for democratic and civil-liberties debates.", "The article treats Palantir as a platform that structures what institutions can see, infer, and act on.", "It provides critical vocabulary for discussing surveillance and asymmetry in data platforms.", "Use for governance critique and theory, not product-level implementation detail." ], "ontology_relevance": "Provides critical academic context for operational data/ontology platforms as surveillance infrastructure.", "ai_relevance": "Relevant to AI-enabled decision systems and governance of analytics platforms.", "palantir_relevance": "Peer-reviewed critical source on Palantir's societal role.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "academic", "civil-liberties", "governance", "information-society", "openalex", "palantir", "platform-power", "predictive-policing", "surveillance" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "ont-ai-032", "title": "The Symbol Grounding Problem", "authors_or_org": "Stevan Harnad", "year": 1990, "source_type": "paper", "bucket": "academic", "url": "https://www.sciencedirect.com/science/article/pii/0167278990900876", "doi_or_identifier": "10.1016/0167-2789(90)90087-6", "venue_or_site": "Physica D", "abstract_or_summary": "Classic paper asking how symbols can acquire meaning rather than being defined only in terms of other symbols.", "key_claims": [ "Purely symbolic systems face a grounding problem if symbols are only connected to other symbols.", "Meaning requires connections to non-symbolic capacities or the world.", "Grounding remains a central issue for AI understanding." ], "ontology_relevance": "Provides conceptual caution that ontologies need grounding in data, perception, and practice.", "ai_relevance": "Foundational grounding critique relevant to LLM meaning debates.", "palantir_relevance": "Supports argument that operational ontology must connect symbols to real objects and workflows.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "meaning", "philosophy-of-ai", "semantics", "symbol-grounding" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-std-nist-sp1270-bias-ai-2022", "title": "Towards a Standard for Identifying and Managing Bias in Artificial Intelligence", "authors_or_org": "Reva Schwartz, Apostol Vassilev, Kristen K. Greene, Lori Perine, Andrew Burt, Patrick Hall", "year": 2022, "source_type": "technical_report", "bucket": "technical", "url": "https://doi.org/10.6028/NIST.SP.1270", "doi_or_identifier": "10.6028/nist.sp.1270", "venue_or_site": "NIST Special Publication 1270", "abstract_or_summary": "NIST report describing bias in AI as technical, social, and organizational, and proposing approaches to identify, measure, and manage bias across AI lifecycles.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "official_docs", "retrieval_tags": [ "ai-governance", "bias", "documentation", "evaluation", "nist", "risk-management", "sociotechnical" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-uberon-multispecies-anatomy-2012", "title": "Uberon, an integrative multi-species anatomy ontology", "authors_or_org": "Christopher J. Mungall; Carlo Torniai; Georgios V. Gkoutos; Suzanna E. Lewis; Melissa A. Haendel", "year": 2012, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1186/gb-2012-13-1-r5", "doi_or_identifier": "10.1186/gb-2012-13-1-r5", "venue_or_site": "Genome Biology", "abstract_or_summary": "Uberon provides an integrative multi-species anatomy ontology for comparative biology, connecting anatomy terms across species and biological resources.", "key_claims": [ "Multi-species anatomy integration requires shared ontology classes and cross-species mappings.", "Uberon enables comparative biology by normalizing anatomical entities.", "Ontology integration supports cross-database biological reasoning." ], "ontology_relevance": "Strong evidence for ontology as semantic interoperability infrastructure across heterogeneous scientific domains.", "ai_relevance": "Cross-species ontology integration is a strong example of semantic normalization for biological discovery and AI-ready datasets.", "palantir_relevance": "Useful analogy for integrating operational entities across organizations and systems with different schemas.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "anatomy", "biomedical-ontology", "interoperability", "phase7", "scientific-infrastructure", "uberon" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase7-bioregistry-biomedical-entity-identification-2022", "title": "Unifying the identification of biomedical entities with the Bioregistry", "authors_or_org": "Charles Tapley Hoyt; et al.", "year": 2022, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1038/s41597-022-01807-3", "doi_or_identifier": "10.1038/s41597-022-01807-3", "venue_or_site": "Scientific Data", "abstract_or_summary": "Bioregistry unifies prefixes, identifiers, provider metadata, URI patterns, and registry mappings for biomedical entities and resources.", "key_claims": [ "Biomedical data integration requires consistent identifier prefixes and URI patterns.", "Registry mappings reduce ambiguity in entity references across resources.", "Identifier governance is a foundation for machine-actionable knowledge graphs." ], "ontology_relevance": "Strengthens the identity layer of ontology: objects need stable, resolvable identifiers before relations are useful.", "ai_relevance": "Reliable AI retrieval and graph construction depend on stable identifiers, namespace mappings, and resolvable entity references.", "palantir_relevance": "Directly analogous to enterprise object identifiers, dataset IDs, and namespace governance in ontology-backed platforms.", "quality_signal": "peer_reviewed", "retrieval_tags": [ "biomedical-ontology", "bioregistry", "entity-identity", "identifiers", "namespace", "phase7", "scientific-infrastructure" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "phase4-palantir-warp-speed-product-2026", "title": "Warp Speed", "authors_or_org": "Palantir Technologies", "year": 2026, "source_type": "product_page", "bucket": "palantir", "url": "https://www.palantir.com/warpspeed", "doi_or_identifier": null, "venue_or_site": "Palantir", "abstract_or_summary": "Public product page for Warp Speed, Palantir's manufacturing-focused offering that applies ontology and AIP-style operating models to industrial production workflows.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "marketing", "retrieval_tags": [ "commercial-claim", "industrial-ai", "manufacturing", "palantir", "warp-speed" ], "triage_tier": "candidate", "triage_score": 42 }, { "id": "oa-https-doi-org-10-1002-ana-27230", "title": "AI in Neurology: Everything, Everywhere, All at Once Part 3: Surveillance, Synthesis, Simulation, and Systems", "authors_or_org": "Matthew Rizzo", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4408884784", "doi_or_identifier": "10.1002/ana.27230", "venue_or_site": "Annals of Neurology", "abstract_or_summary": "This final part 3 review builds on the practical applications discussed in part 2 and explores how artificial intelligence (AI) is transforming data management, neurological education, and neurological care across large healthcare networks and datasets. The review also highlights AI's role in real-world and synthetic data, digital twins, and innovative clinical trial designs, such as in silico and adaptive trials. The review emphasizes AI's ability to drive continuous improvements in care and discovery through comparative effectiveness research and learning health systems. The global healthcare implications discussed here tie back to earlier discussions on human-AI collaboration and precision care, underscoring the neurological sciences' responsibility to adopt AI advances judiciously, while managing their ethical, economic, and environmental impacts. ANN NEUROL 2025;98:651-667.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1109-access-2021-3074021", "title": "A Bio-Inspired Reaction Against Cyberattacks: AIS-Powered Optimal Countermeasures Selection", "authors_or_org": "Pantaleone Nespoli, Félix Gómez Mármol, Jorge Maestre Vidal", "year": 2021, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W3155896443", "doi_or_identifier": "10.1109/access.2021.3074021", "venue_or_site": "IEEE Access", "abstract_or_summary": "Nowadays, Information and Communication Technology (ICT) infrastructures play a crucial role for human beings, providing essential services at astonishing speed. Nevertheless, such a centrality of those infrastructures attracts the interest of ill-motivated actors that target such infrastructures with cyberattacks that are every day more sophisticated and more disruptive. In this alarming context, selecting the optimal set of countermeasures represents a primary need to react against the appearance of potentially dangerous threats effectively. With the motivation to contribute to develop ing faster and more effective response capabilities against them, the paper at hand introduces a novel cybersecurity reaction methodology based on Artificial Immune Systems (AIS), for which the evolutionary computing paradigm has been adopted. By leveraging the outstanding properties of these bio-inspired techniques, the selected countermeasures to defeat cyberthreats through cloning and mutation phases in an effort to improve the quality of the solution from a quantitative perspective, being able to adjust the risk to which the assets of the protected system are exposed. Exhaustive experiments demonstrate the feasibility of the proposed approach, reducing the risk in a more than acceptable time lapse.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1016-j-neuron-2025-09-011", "title": "A transcriptomic atlas of astrocyte heterogeneity across space and time in mouse and marmoset", "authors_or_org": "Margaret E. Schroeder, Dana McCormack, Lukas R. Metzner, Jinyoung Kang, Kenneth K. W. Li, Eunah Yu, Lisa Melamed, Kirsten Levandowski", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4416432377", "doi_or_identifier": "10.1016/j.neuron.2025.09.011", "venue_or_site": "Neuron", "abstract_or_summary": "", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "ont-ai-005", "title": "Accelerating Knowledge Graph and Ontology Engineering with Large Language Models", "authors_or_org": "Cogan Shimizu; Pascal Hitzler", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2411.09601", "doi_or_identifier": "arxiv:2411.09601", "venue_or_site": "arXiv", "abstract_or_summary": "Position paper arguing that LLMs can accelerate ontology modeling, extension, modification, population, alignment, and entity disambiguation, while modular ontology engineering remains central.", "key_claims": [ "LLMs can accelerate key knowledge graph and ontology engineering tasks.", "Ontology modularity is likely to become more important in LLM-assisted workflows.", "LLM-supported ontology engineering needs careful task decomposition and validation.", "LLMs can accelerate multiple knowledge graph and ontology engineering tasks, but the tasks should be decomposed rather than treated as one broad generation problem.", "Modular ontology design is likely central for scalable LLM-assisted ontology and KG workflows.", "KG and ontology engineering should be treated as a lifecycle where construction, alignment, population, and maintenance interact." ], "ontology_relevance": "Current synthesis of LLM impact on ontology and KG engineering practice.", "ai_relevance": "Shows LLMs as tools for generating and maintaining structured knowledge artifacts.", "palantir_relevance": "Maps directly to AI-assisted enterprise ontology lifecycle management.", "quality_signal": "secondary_source", "retrieval_tags": [ "human-in-the-loop", "knowledge-graph-engineering", "lifecycle", "llm", "modular-ontology", "ontology-engineering" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase6-aclu-palantir-deportation-roundup-2026", "title": "All the Ways Palantir Is Assisting Trump's Abusive Removal Campaign", "authors_or_org": "Sophie Feng; Jay Stanley / ACLU", "year": 2026, "source_type": "commercial_article", "bucket": "commercial", "url": "https://www.aclu.org/news/privacy-technology/palantir-deportation-roundup", "doi_or_identifier": null, "venue_or_site": "American Civil Liberties Union", "abstract_or_summary": "ACLU analysis synthesizing public records, investigative reporting, and agency documentation on Palantir systems used in immigration enforcement, including ELITE, ICM, ImmigrationOS, and AI-enhanced tip processing.", "key_claims": [ "Palantir tools consolidate government and private data into enforcement workflows.", "ELITE reportedly maps and prioritizes targets using advanced analytics.", "ImmigrationOS augments ICM for deportation lifecycle management.", "AI tip processing raises transparency and human oversight concerns." ], "ontology_relevance": "Strong comparator for ontology risk: entity resolution, link analysis, locations, people, events, and action workflows can become enforcement infrastructure.", "ai_relevance": "Covers AI-driven outputs, generative AI tip processing, and ImmigrationOS as public-sector AI governance concerns.", "palantir_relevance": "Direct critique of Palantir public-sector deployment in immigration enforcement.", "quality_signal": "secondary_source", "retrieval_tags": [ "aclu", "civil-liberties", "elite", "immigrationos", "palantir", "phase6", "public-sector", "surveillance-ai" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-1108-imds-10-2023-0778", "title": "Antecedents of big data analytics and artificial intelligence adoption on operational performance: the ChatGPT platform", "authors_or_org": "Chin-Tsu Chen, Shih-Chih Chen, Asif Khan, M. K. Lim, Ming‐Lang Tseng", "year": 2024, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/b23e03e0fe1fc48586e115353a5dfa7d2ace5001", "doi_or_identifier": "10.1108/imds-10-2023-0778", "venue_or_site": "Industrial management & data systems", "abstract_or_summary": "PurposeThis study aims to measure the integrated impact of big data analytics and artificial intelligence (BDA-AI) adoption by using the ChatGPT generative AI online platform as a BDA-AI tool on the operational and environmental performance.Design/methodology/approachThis study considers Taiwanese professionals who engage with ChatGPT; the sample consists of 388 online users.FindingsThis study’s main finding is that the considered antecedents – including technological, organizational and environmental contexts, tangible resources and workforce skills – are significantly associated with BDA-AI adoption. Notably, BDA-AI adoption exhibits a significant relationship with operational performance, environmental performance and environmental process integration. Moreover, environmental process integration is significantly correlated with environmental performance. Lastly, operational performance is significantly correlated with environmental performance.Originality/valueThis study contributes to the heavily lacking but developing literature on the antecedents and consequences of BDA-AI adoption. Its theoretical foundation consists of the technological-organizational-environmental model, Roger’s diffusion of innovation theory and resource-based view theory.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "industry-commercial-cambridge-semantics-anzo", "title": "Anzo knowledge graph data fabric materials", "authors_or_org": "Cambridge Semantics", "year": 2026, "source_type": "commercial_article", "bucket": "commercial", "url": "https://www.cambridgesemantics.com/product/anzo", "doi_or_identifier": null, "venue_or_site": "Cambridge Semantics", "abstract_or_summary": "Anzo materials frame enterprise knowledge graphs as a data fabric layer for integrating siloed data, harmonizing entities, and enabling analytics over semantic models.", "key_claims": [ "Semantic models can harmonize heterogeneous enterprise sources into business-facing entities and relationships.", "Virtualization and graph integration can reduce the need for monolithic data consolidation.", "Knowledge graphs can support analytics in regulated and complex domains." ], "ontology_relevance": "Strong commercial example of ontology as data fabric and semantic harmonization layer.", "ai_relevance": "Provides structured context that can support AI search, analytics, and retrieval workflows.", "palantir_relevance": "Relevant comparator to Palantir Foundry's integration and ontology layer.", "quality_signal": "marketing", "retrieval_tags": [ "anzo", "cambridge-semantics", "data-fabric", "enterprise-kg", "semantic-layer", "virtualization" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-59075-jssa-v3i2-265", "title": "Artificial Intelligence in E-Commerce: A Comparative Study of AI-Driven Innovations in Amazon and Daraz within the Pakistani Market", "authors_or_org": "Madiha Tariq, Ali Fayyaz Munir, Ali Raza Elahi, Zahra Zainab, Maria Naqash, Anam Fatima", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/49e46379db1bc7795f54da95146095e592248a33", "doi_or_identifier": "10.59075/jssa.v3i2.265", "venue_or_site": "Journal for Social Science Archives", "abstract_or_summary": "The study investigates the strong influence of artificial intelligence on reshaping the e-commerce platform in Pakistan. The global e-commerce industries use several artificial techniques to promote their business for profitability. This research paper explores a comparative analysis of the global leading e-commerce website Amazon and the local South Asian website Daraz. The study also attempts to analyze the level of AI utilization tools in customer experience, streamlining business operations, fraud detection, and supply chain management. By using convenience sampling techniques, 400 questionnaires were distributed, out of which 345 were received back. Only 315 responses were deemed valid for analysis. The study analyzed the data analyzed with SmartPLS. The study findings underscore that the strategic deployment of AI tools across customer engagement, operational streamlining, fraud mitigation, and supply chain optimization have a profound and significant impact on Daraz’s e-commerce performance. Furthermore, this research also recommended that Daraz enhance the usage level of artificial intelligence techniques in their business activities, transforming the business model to increase profitability and take a step into a global e-commerce platform.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-1108-ijoa-09-2023-3992", "title": "Artificial intelligence in talent acquisition: exploring organisational and operational dimensions", "authors_or_org": "Dhyana Paramita, Simon Okwir, Cali Nuur", "year": 2024, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/9d6c28d72ee9cd4471e556c94038856fd5d27e13", "doi_or_identifier": "10.1108/ijoa-09-2023-3992", "venue_or_site": "The International Journal of Organizational Analysis", "abstract_or_summary": "\nPurpose\nWith the recent proliferation of AI, organisations are transforming not only their organisational design but also the input and output operational processes of the hiring process. The purpose of this paper is to explore the organisational and operational dimensions resulting from the deployment of AI during talent acquisition process.\n\n\nDesign/methodology/approach\nThe authors conducted semi-structured interviews and meetings with human resources (HRs) professionals, recruiters and AI hiring platform providers in Sweden. Using an inductive data analysis rooted in the principles of grounded theory, the study uncovered four aggregate dimensions critical to understanding the role of AI in talent acquisition.\n\n\nFindings\nWith insights from algorithmic management and ambidexterity theory, the study presents a comprehensive theoretical framework that highlights four aggregate dimensions describing AI’s transformative role in talent recruitment. The results provide a cautionary perspective, advising against an excessive emphasis on operational performance driven solely by algorithmic management.\n\n\nResearch limitations/implications\nThe study is limited in scope and subject to several constraints. Firstly, the sample size and diversity are restricted, as the findings are based on a limited number of semi-structured interviews and meetings with HRs professionals, recruiters, and AI hiring platform providers. Secondly, the rapid evolution of AI technologies means that the study’s findings may quickly become outdated as new advancements and applications emerge.\n\n\nPractical implications\nThe results provide managers with actionable information that can lead to more precise and strategic management practices, ultimately contributing to improved organizational performance and outcomes. Plus, enhancing their ability to make informed decisions, optimize processes and address challenges effectively.\n\n\nSocial implications\nThe results signal both positive and negative impacts on employment opportunities. On the positive side, AI can streamline recruitment processes, making it easier for qualified candidates to be identified and hired quickly. However, AI systems can also perpetuate existing biases present in the data they are trained on, leading to unfair hiring practices where certain groups are systematically disadvantaged.\n\n\nOriginality/value\nBy examining the balance between transactional efficiency and relational engagement, the research addresses a crucial trade-off that organizations face when implementing AI in recruitment. The originality lies in its critique of the prevailing emphasis on e-recruiting.\n", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-36647-ciml-06-02-a006", "title": "Artificial Intelligence-Based Time Series Forecasting for Operational Risk Management in Healthcare: A Comparative Study of ARIMA, SARIMA, and Prophet on the EPA Platform", "authors_or_org": "Mateus Marinho, Isadora Stéfany Rezende Remigio Mesquita, Lucas Elias Cardoso Rocha, D. A. Vieira, T. Kudo, R. Braga", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/ca0cf5b9d83bb6eeeae3f82121600a1839e4f963", "doi_or_identifier": "10.36647/ciml/06.02.a006", "venue_or_site": "Computational Intelligence and Machine Learning", "abstract_or_summary": "This study presents the development and evaluation of time series forecasting models for operational risk management in healthcare institutions, using real data from the EPA Platform. The objective was to predict the number of risk-related occurrences based on historical records, comparing the performance of ARIMA, SARIMA, and Prophet. The models were applied to both daily and monthly granularities, which enabled the identification of the most effective forecasting strategies for each temporal context. The approach included seasonal pattern analysis and hyperparameter optimization through Grid Search. Experimental results highlighted each model's ability to capture complex temporal dynamics while maintaining low computational cost. The findings supported strategic decision-making aimed at risk mitigation and patient safety, and demonstrated the practical applicability of AI-based forecasting in institutional healthcare environments.\n\nKeyword :Artificial Intelligence, Time Series Forecasting, ARIMA, SARIMA, Prophet, Healthcare Risk Management", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1093-police-paad061", "title": "Big data policing: The use of big data and algorithms by the Netherlands Police", "authors_or_org": "M.B. Schuilenburg, Melvin Soudijn", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4387031484", "doi_or_identifier": "10.1093/police/paad061", "venue_or_site": "Policing A Journal of Policy and Practice", "abstract_or_summary": "Abstract In recent years, the rise of big data has revolutionized many domains, including policing. Research is lacking, however, on the various ways in which the police use big data applications. This study provides new insights into the ways the Netherlands Police currently use big data and algorithmic applications. Based on a novel data source—job vacancies in the IT domain for the Netherlands Police—we distinguish three areas in which big data is used: frontline policing, criminal investigations, and intelligence. Our research shows that the use of big data by the Netherlands Police mainly involves relatively simple applications and that—in contrast to police forces in the USA—big data applications with the objective of assessing risks are the least common. The research also shows that big data policing leads to greater discretionary powers for police functions such as software developers and network designers.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-48550-arxiv-2304-14975", "title": "Concept-centric Software Development", "authors_or_org": "Peter Wilczynski, Taylor Gregoire-Wright, Daniel J. Jackson", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4367628396", "doi_or_identifier": "10.48550/arxiv.2304.14975", "venue_or_site": "arXiv (Cornell University)", "abstract_or_summary": "Developers have long recognized the importance of the concepts underlying the systems they build, and the primary role concepts play in shaping user experience. To date, however, concepts have tended to be only implicit in software design with development being organized instead around more concrete artifacts (such as wireframes and code modules). Palantir, a software company whose data analytics products are widely used by major corporations, recently reworked its internal representation of its software development process to bring concepts to the fore, making explicit the concepts underlying its products, how they are clustered, used within and across applications, and governed by teams. With a centralized repository of concepts, Palantir engineers are able to align products more closely based on shared concepts, evolve concepts in response to user needs, and communicate more effectively with non-engineering groups within the company. This paper reports on Palantir's experiences to date, analyzing both successes and challenges, and offers advice to other organizations considering adopting a concept-centric approach to software development", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1145-3622758-3622894", "title": "Concept-Centric Software Development: An Experience Report", "authors_or_org": "Peter Wilczynski, Taylor Gregoire-Wright, Daniel J. Jackson", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4387773528", "doi_or_identifier": "10.1145/3622758.3622894", "venue_or_site": "", "abstract_or_summary": "Developers have long recognized the importance of the concepts underlying the systems they build, and the primary role that concepts play in shaping user experience. To date, however, concepts have tended to be only implicit in software design with development being organized instead around more concrete artifacts (such as wireframes and code modules).", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1080-1369118x-2024-2442394", "title": "Data integration and analysis platforms as digital platforms: a conceptual proposal", "authors_or_org": "Simon Egbert, Lena Ulbricht", "year": 2024, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4405691926", "doi_or_identifier": "10.1080/1369118x.2024.2442394", "venue_or_site": "Information Communication & Society", "abstract_or_summary": "In this paper, we conceptualize platforms that aim at data integration and analysis as a distinct type of digital platform. Based on the existing literature on digital platforms in general, which so far did not engage with data integration and analysis platforms, we present a definition of data integration and analysis platforms as a digital platform type in its own right. To emphasize the specificity of this platform type, we present a case example, Palantir Technologies, and highlight its structural characteristics and key technical features. The case study shows that data integration and analysis services should be understood as platforms, because like other platforms, they serve as modifiable digital infrastructures that bring together different parties and enable data-dependent interaction. It is especially important to understand data integration and analysis platforms as digital platforms because these platforms have their own politics and determine what happens on them, as we illustrate with regard to two important social values that are part of the data sovereignty of organizations: epistemic opacity and epistemic control. We conclude that it is time that platform research not only scrutinizes those technologies and companies that are visible to the eye of large numbers of end-users, but also those that operate in the dark.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "industry-commercial-dataworld-llm-benchmark", "title": "data.world materials on knowledge graphs and LLM accuracy benchmarks", "authors_or_org": "data.world", "year": 2024, "source_type": "commercial_article", "bucket": "commercial", "url": "https://data.world/resources", "doi_or_identifier": null, "venue_or_site": "data.world Resources", "abstract_or_summary": "data.world resource materials include claims that knowledge graph context improves LLM or natural-language analytics accuracy, framed around its AI Context Engine and catalog graph.", "key_claims": [ "Knowledge graph context is claimed to improve LLM answers over enterprise data.", "The reusable architecture is governed metadata plus business terms plus lineage as AI context.", "Benchmark numbers should be treated as vendor claims unless the full experimental setup is independently evaluated." ], "ontology_relevance": "Useful for catalog-backed ontology as AI context.", "ai_relevance": "Directly relevant to LLM grounding and enterprise AI accuracy claims.", "palantir_relevance": "Commercial comparator to Palantir AIP claims about ontology-grounded AI.", "quality_signal": "marketing", "retrieval_tags": [ "accuracy-claim", "ai-context-engine", "catalog", "data-world", "knowledge-graph", "llm-benchmark" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1007-s00146-025-02678-z", "title": "Decider, overruler, or trusted companion? An exploration of the advent of the virtual human twin and its impact on decision-making in healthcare", "authors_or_org": "Karen Sonego, Thomas Stead", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4415566818", "doi_or_identifier": "10.1007/s00146-025-02678-z", "venue_or_site": "AI & Society", "abstract_or_summary": "In recent years, the healthcare sector has seen growing interest in the utility and acceptability of virtual human twins. Data-intensive, computational model-based tools are crucial for decision support and risk prevention in healthcare, with the potential to autonomously analyse scenarios and generate predictive insights to inform decision-making. The concept of the virtual human twin, also referred to as a human digital twin or simply a digital twin, holds the potential to offer this capability, paving the way for personalised, data-driven healthcare.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase6-ico-dhsc-fdp-foi-2026", "title": "Department of Health & Social Care Decision Notice IC-413447-N9X2", "authors_or_org": "Information Commissioner's Office", "year": 2026, "source_type": "webpage", "bucket": "technical", "url": "https://ico.org.uk/action-weve-taken/decision-notices/2026/03/ic-413447-n9x2", "doi_or_identifier": "ic-413447-n9x2", "venue_or_site": "Information Commissioner's Office", "abstract_or_summary": "ICO decision notice concerning requested information about the NHS FDP contract with Palantir Technologies Ltd. The ICO accepted DHSC's reliance on FOIA section 35(1)(a) for policy formulation/development.", "key_claims": [ "The requested information concerned the NHS FDP contract with Palantir.", "The ICO found that the section 35(1)(a) exemption applied.", "No further steps were required." ], "ontology_relevance": "Useful for mapping transparency limits around infrastructure decisions that define data models and platform control.", "ai_relevance": "Shows disclosure constraints around a major AI-adjacent public data platform.", "palantir_relevance": "Directly concerns Palantir FDP contract transparency.", "quality_signal": "primary_source", "retrieval_tags": [ "dhsc", "fdp", "foi", "ico", "palantir", "phase6", "transparency" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1080-1369118x-2024-2371803", "title": "Desiloization and its discontents: the politics of data storage in the age of platformization", "authors_or_org": "Nanna Bonde Thylstrup, Matthew Archer, Henriette Steiner", "year": 2024, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4400591355", "doi_or_identifier": "10.1080/1369118x.2024.2371803", "venue_or_site": "Information Communication & Society", "abstract_or_summary": "This essay explores how desiloization strategies facilitate the emergence and increasing centralization of private platform power, rather than flattening and democratizing access to and control over data. Drawing on both an original genealogy of desiloization discourses and a case study of the World Bank’s evolving approach to desiloization, we show how the slipperiness of the notion of ‘silos’ obscures the way desiloization efforts reinforce the structural power of ‘Big Tech,’ a key characteristic of the emerging paradigm of platform capitalism. More specifically, we argue that concerns about interoperability often elide its reliance on the products and services of private, for-profit companies, engendering a mode of corporate centralization. In doing so, we build on an emergent interdisciplinary understanding of storage as an ‘infrastructural ecology,’ shifting attention from the biophysical environments to the political environments in which discourses and strategies of desiloization circulate and transpire.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-5194-isprs-annals-x-m-2-2025-39-2025", "title": "Development of an Artificial Intelligence-based Platform for the Analysis and Utilization of Cultural Heritage Data", "authors_or_org": "S. Baek, Hyerin Hwang, Chan-Woo Park, Hee-Kwon Kim, Jae-Ho Lee", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/69414c998d85eaf1def06be9a53f715349811962", "doi_or_identifier": "10.5194/isprs-annals-x-m-2-2025-39-2025", "venue_or_site": "ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences", "abstract_or_summary": "Abstract. This research presents the development of an AI-powered digital cultural heritage platform designed to address the growing need for efficient management, analysis, and utilization of traditional cultural assets in the era of rapid digital transformation. The platform integrates a digital archive, data-driven AI analysis modules, generative AI techniques for content enrichment, and an operational environment for demonstration and deployment. It supports the stable storage and intelligent processing of diverse cultural heritage data. A notable feature is the implementation of an AI-based relational analysis model that captures the complex metadata and structural relationships inherent in cultural heritage objects, enabling the automatic identification and visualization of meaningful semantic connections. The platform also incorporates dedicated viewers—such as RTI, Giga Pixel, GLB, and NXG—offering intuitive access to ultra-high-resolution and 3D representations of cultural objects. This multifaceted system supports a wide range of applications in education, academic research, and exhibition contexts, demonstrating its versatility and strong potential for practical implementation across public and scholarly domains.\n", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1007-978-3-030-02330-0", "title": "Disrupting Finance", "authors_or_org": "Theo Lynn, John G. Mooney, Pierangelo Rosati, Mark Cummins", "year": 2018, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W2943467878", "doi_or_identifier": "10.1007/978-3-030-02330-0", "venue_or_site": "Palgrave studies in digital business & enabling technologies", "abstract_or_summary": "This open access book defines a fintech ecosystem for the 21st century, considering a range of technologies.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-bernstein-2011-schema-matching-decade", "title": "Generic Schema Matching, Ten Years Later", "authors_or_org": "Philip A. Bernstein, Jayant Madhavan, Erhard Rahm", "year": 2011, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.14778/2002938.2002940", "doi_or_identifier": "10.14778/2002938.2002940", "venue_or_site": "Proceedings of the VLDB Endowment", "abstract_or_summary": "Reviews progress and open problems in schema matching, including reusable matchers, user involvement, evaluation, and integration workflows.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "data-integration", "evaluation", "interoperability", "schema-matching" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-jin-2024-llms-on-graphs", "title": "Large Language Models on Graphs: A Comprehensive Survey", "authors_or_org": "Bowen Jin, Gang Liu, Chi Han, Meng Jiang, Heng Ji, Jiawei Han", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://arxiv.org/abs/2312.02783", "doi_or_identifier": "10.48550/arxiv.2312.02783", "venue_or_site": "IEEE Transactions on Knowledge and Data Engineering", "abstract_or_summary": "Reviews how language models are applied to pure graphs, text-attributed graphs, and text-paired graphs, including prediction, encoding, alignment, and graph reasoning.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "graph-learning", "graphs", "knowledge-graph", "llm", "survey" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-bizer-2009-linked-data", "title": "Linked Data - The Story So Far", "authors_or_org": "Christian Bizer, Tom Heath, Tim Berners-Lee", "year": 2009, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.4018/jswis.2009081901", "doi_or_identifier": "10.4018/jswis.2009081901", "venue_or_site": "International Journal on Semantic Web and Information Systems", "abstract_or_summary": "Explains Linked Data principles and early deployment patterns for publishing and connecting structured data on the web.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "interoperability", "linked-data", "semantic-web", "web-of-data" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "ont-ai-029", "title": "Neuro-Symbolic Artificial Intelligence: The State of the Art", "authors_or_org": "A. S. d'Avila Garcez; L. C. Lamb; D. M. Gabbay; D. S. Weld; P. Hitzler; L. Serafini; M. Spranger; editors including Pascal Hitzler and Md Kamruzzaman Sarker", "year": 2022, "source_type": "book", "bucket": "books", "url": "https://ebooks.iospress.nl/volume/neuro-symbolic-artificial-intelligence-the-state-of-the-art", "doi_or_identifier": "isbn:978-1-64368-244-2", "venue_or_site": "IOS Press", "abstract_or_summary": "Edited volume surveying neuro-symbolic AI methods, applications, and open problems across logic, learning, knowledge representation, and reasoning.", "key_claims": [ "Hybrid neural-symbolic systems are motivated by explainability, reasoning, and data efficiency.", "Knowledge representation remains important in modern AI architectures.", "The field includes multiple integration strategies rather than one dominant architecture." ], "ontology_relevance": "Connects formal knowledge representation to modern AI architectures.", "ai_relevance": "Major state-of-the-art synthesis for neuro-symbolic AI.", "palantir_relevance": "Useful for framing enterprise ontology as a neuro-symbolic system component.", "quality_signal": "secondary_source", "retrieval_tags": [ "knowledge-representation", "neuro-symbolic-ai", "reasoning", "state-of-art" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-bommasani-2021-foundation-models", "title": "On the Opportunities and Risks of Foundation Models", "authors_or_org": "Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, Shyamal Buch, Dallas Card, Rodrigo Castellon, Niladri Chatterji, Ananth Chen, Kathleen Creel, Jared Quincy Davis, Dora Demszky, Chris Donahue, Moussa Doumbouya, Esin Durmus, Stefano Ermon, John Etchemendy, Kawin Ethayarajh, Li Fei-Fei, Chelsea Finn, Trevor Gale, Lauren Gillespie, Karan Goel, Noah Goodman, Shelby Grossman, Neel Guha, Tatsunori Hashimoto, Peter Henderson, John Hewitt, Daniel E. Ho, Jenny Hong, Kyle Hsu, Jing Huang, Thomas Icard, Saahil Jain, Dan Jurafsky, Pratyusha Kalluri, Siddharth Karamcheti, Geoff Keeling, Fereshte Khani, Omar Khattab, Pang Wei Koh, Mark Krass, Ranjay Krishna, Rohith Kuditipudi, Ananya Kumar, Faisal Ladhak, Mina Lee, Tony Lee, Jure Leskovec, Isabelle Levent, Xiang Lisa Li, Xuechen Li, Tengyu Ma, Ali Malik, Christopher D. Manning, Suvir Mirchandani, Eric Mitchell, Zanele Munyikwa, Suraj Nair, Avanika Narayan, Deepak Narayanan, Ben Newman, Allen Nie, Juan Carlos Niebles, Hamed Nilforoshan, Julian Nyarko, Girish Ogut, Laurel Orr, Isabel Papadimitriou, Joon Sung Park, Chris Piech, Eva Portelance, Christopher Potts, Aditi Raghunathan, Rob Reich, Hongyu Ren, Frieda Rong, Yusuf Roohani, Camilo Ruiz, Jack Ryan, Christopher Ré, Dorsa Sadigh, Shiori Sagawa, Keshav Santhanam, Andy Shih, Krishnan Srinivasan, Alex Tamkin, Rohan Taori, Armin W. Thomas, Florian Tramèr, Rose E. Wang, William Wang, Bohan Wu, Jiajun Wu, Yuhuai Wu, Sang Michael Xie, Michihiro Yasunaga, Jiaxuan You, Matei Zaharia, Michael Zhang, Tianyi Zhang, Xikun Zhang, Yuhui Zhang, Lucia Zheng, Kaitlyn Zhou, Percy Liang", "year": 2021, "source_type": "report", "bucket": "academic", "url": "https://arxiv.org/abs/2108.07258", "doi_or_identifier": "10.48550/arxiv.2108.07258", "venue_or_site": "arXiv / Stanford CRFM", "abstract_or_summary": "Comprehensive report on foundation model capabilities, risks, data issues, evaluation, societal impacts, and governance needs.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "evaluation", "foundation-models", "governance", "llm", "risks" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "sep_ontology_is_2026", "title": "Ontology and Information Systems", "authors_or_org": "Fabian Neuhaus and Amanda Vizedom, Stanford Encyclopedia of Philosophy", "year": 2026, "source_type": "webpage", "bucket": "academic", "url": "https://plato.stanford.edu/entries/ontology-is", "doi_or_identifier": "sep entry, first published 2026-01-03", "venue_or_site": "Stanford Encyclopedia of Philosophy", "abstract_or_summary": "Current reference article explaining ontology in information systems, including shared meaning, domain and upper ontologies, ontological commitment, logical languages, assessment, applications, and large language models.", "key_claims": [ "Ontology in information systems creates shared meanings of symbols for computational use.", "Ontologies can support communication among people, communication among machines, and computation.", "Concept drift and interoperability are central motivations for ontology engineering.", "The usefulness of an ontology depends on its intended use and level of formal commitment." ], "ontology_relevance": "Provides a contemporary, authoritative bridge between philosophical ontology and computational ontology practice.", "ai_relevance": "Directly frames ontology as a response to concept drift, interoperability, and LLM-era grounding problems.", "palantir_relevance": "Useful lens for evaluating operational ontology as shared meaning across humans, software systems, and AI workflows.", "quality_signal": "secondary_source", "retrieval_tags": [ "concept-drift", "information-systems", "llm", "ontological-commitment", "ontology", "upper-ontology" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "ont-ai-026", "title": "OntoRAG: Ontology-Based Retrieval Augmented Generation for Interpretable and Explainable Answers", "authors_or_org": "Nemanja Brankovic; Georg Lausen", "year": 2024, "source_type": "paper", "bucket": "academic", "url": "https://openreview.net/forum?id=Yh1c0wY2vH", "doi_or_identifier": "openreview:yh1c0wy2vh", "venue_or_site": "OpenReview", "abstract_or_summary": "Proposes an ontology-based RAG approach intended to improve interpretability and explainability through ontology-guided retrieval.", "key_claims": [ "Ontology-guided retrieval can improve answer transparency.", "Structured concepts and relations can provide explanations for retrieved context.", "Evaluation of ontology-RAG must consider both answer quality and explanation quality.", "Structured concepts and relations can provide explanation paths for retrieved context.", "Evaluation should include explanation quality, not only answer accuracy." ], "ontology_relevance": "Explicitly frames ontology as the retrieval and explanation scaffold for RAG.", "ai_relevance": "Recent LLM retrieval architecture focusing on interpretability.", "palantir_relevance": "Relevant to explaining answers over governed enterprise ontology objects.", "quality_signal": "secondary_source", "retrieval_tags": [ "concept-paths", "explainability", "llm", "ontology-rag", "ontorag", "structured-retrieval" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "guardian-palantir-nhs-access-2026", "title": "Palantir's access to identifiable NHS England patient data is 'dangerous', MPs say", "authors_or_org": "The Guardian", "year": 2026, "source_type": "commercial_article", "bucket": "commercial", "url": "https://www.theguardian.com/society/2026/may/11/palantir-access-nhs-england-patient-data", "doi_or_identifier": null, "venue_or_site": "The Guardian", "abstract_or_summary": "News report on concerns from MPs and advocacy groups about Palantir access to identifiable NHS patient data as part of the Federated Data Platform program, with NHS/Palantir responses.", "key_claims": [ "MPs and advocacy groups warned that Palantir access to identifiable patient data could damage public trust.", "The report says NHS argued access was limited, monitored, and subject to clearance.", "Palantir is described as acting as a data processor under NHS direction.", "MPs and critics expressed concern about identifiable patient data and Palantir's role.", "The report is useful for tracking public controversy but should be checked against official NHS and parliamentary materials.", "It reflects reputational and legitimacy risks in public-sector operational data platforms.", "MPs and advocacy groups raised concerns about Palantir access to identifiable NHS patient data.", "The controversy centers on consent, transparency, and public trust.", "NHS and Palantir positions frame access as limited, monitored, and processor-bound." ], "ontology_relevance": "Highlights governance risk when operational data platforms include sensitive person-level data.", "ai_relevance": "Relevant to AI infrastructure trust, data minimization, and accountability.", "palantir_relevance": "Current third-party reporting on a major Palantir public-sector controversy.", "quality_signal": "secondary_source", "retrieval_tags": [ "guardian", "identifiable-data", "nhs", "palantir", "patient-data", "privacy", "public-consent", "public-sector", "public-trust" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "bmj-palantir-nhs-coalition-2026", "title": "Palantir: Coalition urges NHS organisations to refuse to comply with data platform directive", "authors_or_org": "The BMJ", "year": 2026, "source_type": "commercial_article", "bucket": "technical", "url": "https://www.bmj.com/content/392/bmj.s481", "doi_or_identifier": "bmj 2026;392:s481", "venue_or_site": "The BMJ", "abstract_or_summary": "BMJ news coverage of a coalition urging NHS organisations to refuse use of software operated by Palantir, and reporting Palantir's rejection of some claims.", "key_claims": [ "A coalition urged NHS organisations to resist a directive involving Palantir-operated FDP software.", "Palantir rejected claims including claims about engineers accessing patient identifiable data.", "The article reflects ongoing healthcare-sector trust and governance disputes.", "Civil-society and professional actors continued to raise privacy concerns about the FDP in 2026.", "The article is useful evidence of persistent governance controversy after contract award.", "Claims should be triangulated with NHS primary sources and parliamentary debate." ], "ontology_relevance": "Shows societal and institutional resistance to ontology/data-platform centralization in healthcare.", "ai_relevance": "Relevant to AI deployment legitimacy and data trust in clinical environments.", "palantir_relevance": "Independent medical-news source on Palantir NHS controversy.", "quality_signal": "secondary_source", "retrieval_tags": [ "bmj", "campaigners", "critique", "fdp", "governance", "health-data", "nhs", "palantir", "privacy" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-openalex-org-w7162712905", "title": "Reality Constructed, Reality Lost: Navigating Multiple Digital Realities Of Sustainability", "authors_or_org": "Jingyang Wang, Mathias Muench, Elizabeth A. Teracino, Christine Legner, Joerg Hans Mayer, Reiner Quick", "year": 2026, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W7162712905", "doi_or_identifier": null, "venue_or_site": "Journal of the Association for Information Systems", "abstract_or_summary": "Organizations require data to assess their sustainability commitments. While research calls for a data perspective on sustainability, it remains underexplored how data represent and mediate these commitments. In this paper, we argue that data function as semiotic artifacts deeply imbricated in organizational contexts that inform and constrain real-world sustainability efforts. Based on a revelatory case study, we identify three types of digital realities related to sustainability: (1) an operational reality, grounded in time-bound records of physical events; (2) a strategic reality, constructed through sustainability reporting; (3) a bridging reality, codified through quality standards and control workflows. We thereby show how multiple, co-existing digital realities are constructed and aligned within the sustainability domain. Theoretically, our study extends representation theory by shifting the focus from how individual representations mediate reality to how multiple, co-existing digital realities are constructed, interact, and must be aligned within organizations.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase3-sam-n3c-palantir-contract-2025", "title": "Secure Platforms Support for the N3C Data Enclave", "authors_or_org": "U.S. General Services Administration / SAM.gov", "year": 2025, "source_type": "webpage", "bucket": "synthesis", "url": "https://sam.gov/opp/0b0ee7b24ab8472e856994709c2ac9dc/view", "doi_or_identifier": null, "venue_or_site": "SAM.gov", "abstract_or_summary": "U.S. federal procurement notice describing secure cloud platform-as-a-service support for the N3C Data Enclave, a public record relevant to Palantir's role in health-data platform operations.", "key_claims": [ "The procurement concerns secure cloud PaaS support for the N3C Data Enclave.", "N3C is described as a national clinical cohort collaborative data enclave.", "Procurement records provide contractual context separate from vendor marketing." ], "ontology_relevance": "Shows public procurement context for operational data-platform infrastructure.", "ai_relevance": "Relevant to government sourcing of AI-ready data environments.", "palantir_relevance": "Public procurement evidence for Palantir/N3C platform support context.", "quality_signal": "primary_source", "retrieval_tags": [ "data-enclave", "n3c", "palantir", "procurement", "public-contract", "sam.gov" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-ji-2023-hallucination-survey", "title": "Survey of Hallucination in Natural Language Generation", "authors_or_org": "Ziwei Ji, Nayeon Lee, Rita Frieske, Tiezheng Yu, Dan Su, Yan Xu, Etsuko Ishii, Ye Jin Bang, Andrea Madotto, Pascale Fung", "year": 2023, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1145/3571730", "doi_or_identifier": "10.1145/3571730", "venue_or_site": "ACM Computing Surveys", "abstract_or_summary": "Surveys hallucination definitions, causes, detection, and mitigation across natural language generation, showing why fluent text alone is not reliable knowledge infrastructure.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "evaluation", "hallucination", "llm-limitations", "natural-language-generation", "survey" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-openalex-org-w7164780573", "title": "The Automation of War: What Palantir Reveals About Modern Military Ethics", "authors_or_org": "Johnny Mou", "year": 2026, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W7164780573", "doi_or_identifier": null, "venue_or_site": "eCommons (Cornell University)", "abstract_or_summary": "", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-mittelstadt-2016-ethics-algorithms", "title": "The Ethics of Algorithms: Mapping the Debate", "authors_or_org": "Brent Daniel Mittelstadt, Patrick Allo, Mariarosaria Taddeo, Sandra Wachter, Luciano Floridi", "year": 2016, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1177/2053951716679679", "doi_or_identifier": "10.1177/2053951716679679", "venue_or_site": "Big Data & Society", "abstract_or_summary": "Maps ethical concerns around algorithmic systems, including opacity, bias, responsibility gaps, and evidence for decision-making.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "algorithmic-accountability", "ethics", "governance", "opacity" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "phase4-nature-shadbolt-2006-semantic-web-revisited", "title": "The Semantic Web Revisited", "authors_or_org": "Nigel Shadbolt, Wendy Hall, Tim Berners-Lee", "year": 2006, "source_type": "paper", "bucket": "academic", "url": "https://doi.org/10.1109/MIS.2006.62", "doi_or_identifier": "10.1109/mis.2006.62", "venue_or_site": "IEEE Intelligent Systems", "abstract_or_summary": "Revisits the Semantic Web vision after early deployment experience, clarifying practical progress, misconceptions, and the need for linked, reusable data.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "peer_reviewed_survey", "retrieval_tags": [ "history", "infrastructure", "linked-data", "semantic-web" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "s2-10-1109-iot-siu65919-2025-11402773", "title": "The Smart Industrial Transformation: The Integration of Big Data, Artificial Intelligence (AI), Internet of Things (IoT), and Virtual Reality (VR) Technologies", "authors_or_org": "Farman Ali, Neha Singh, Sanjay Singh Chauhan", "year": 2025, "source_type": "paper", "bucket": "palantir", "url": "https://www.semanticscholar.org/paper/634ef307020c251d8519e878a7b59229d87bb9ea", "doi_or_identifier": "10.1109/iot-siu65919.2025.11402773", "venue_or_site": "International Conference on Internet of Things", "abstract_or_summary": "The Internet of Things (IoT) is a technology which helps in supporting communication between physical and digital entities via mobile connectivity. The main reason for using Internet computing is to streamline human being work and improve the individual experience and operational accessibility. This is a very reliable platform for creating advanced Industry that will be giving 5.0 services and applications. The research shows the in-depth study of Scopus listed articles where IOT has been used to benefit the consumers and to understand the outcome. The main tools used in the Internet of Things (IoT) includes, encompassing big data, cloud computing, augmented reality, virtual reality, and radio frequency identification. The review article mainly focused on key factors of IOT and how it has affected the manufacturing of the products, Architecture or preparing business models. Finally, we have made the study specific to some potential uses of IoT and artificial intelligence in Industry 5.0 which will benefit the producer and consumer both. Although the smart tool of Industry 5.0 are more used in small and medium size industries but it is also making some presence in large scale companies and benefiting them too.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "palantir", "semantic-scholar" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1093-ips-olab003", "title": "Toward a Critique of Algorithmic Violence", "authors_or_org": "Rocco Bellanova, Kristina Irion, Katja Lindskov Jacobsen, Francesco Ragazzi, Rune Andersen, Lucy Suchman", "year": 2021, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W3139232402", "doi_or_identifier": "10.1093/ips/olab003", "venue_or_site": "International Political Sociology", "abstract_or_summary": "Abstract Questions about how algorithms contribute to (in)security are under discussion across international political sociology. Building upon and adding to these debates, our collective discussion foregrounds questions about algorithmic violence. We argue that it is important to examine how algorithmic systems feed (into) specific forms of violence, and how they justify violent actions or redefine what forms of violence are deemed legitimate. Bringing together different disciplinary and conceptual vantage points, this collective discussion opens a conversation about algorithmic violence focusing both on its specific instances and on the challenges that arise in conceptualizing and studying it. Overall, the discussion converges on three areas of concern—the violence undergirding the creation and feeding of data infrastructures; the translation processes at play in the use of computer/machine vision across diverse security practices; and the institutional governing of algorithmic violence, especially its organization, limitation, and legitimation. Our two-fold aim is to show the potential of a cross-disciplinary conversation and to move toward an interactional research agenda. While our approaches diverge, they also enrich each other. Ultimately, we highlight the critical purchase of studying the role of algorithmic violence in the fabric of the international through a situated analysis of algorithmic systems as part of complex, and often messy, practices. Les questions concernant la manière dont les algorithmes affectent l’(in)sécurité deviennent de plus en plus courantes en sociologie politique internationale. Notre discussion collective s'appuie sur ces débats et les enrichit en abordant les questions portant sur la violence algorithmique. Nous soutenons qu'il est important d'analyser et de discuter de la manière dont les systèmes algorithmiques alimentent (et entretiennent) des formes spécifiques de violence, ainsi que de la façon dont ils justifient des actes violents ou redéfinissent les formes de violence jugées légitimes. Cette discussion collective réunit différents points de vue disciplinaires et conceptuels pour ouvrir un débat sur la violence algorithmique en se concentrant à la fois sur des exemples spécifiques et sur les défis à relever pour la conceptualiser et l’étudier. Cette discussion se concentre sur trois sujets de préoccupation : la violence qui sous-tend la création et l'alimentation des infrastructures de données, les processus de conversion en jeu dans l'utilisation de la vision informatique/machine à travers diverses pratiques de sécurité, et la gouvernance institutionnelle de la violence algorithmique, en particulier son organisation, sa limitation et sa légitimation. Notre double objectif est de montrer le potentiel d'une discussion interdisciplinaire et d'avancer vers un programme de recherche interactionnel. Bien que nos approches divergent, elles s'enrichissent mutuellement. Notre but est de mettre en évidence les possibilités analytiques ouvertes par l'étude de la violence algorithmique et de son role dans la fabrique des relations internationales, par le biais d'une étude des systèmes algorithmiques dans le cadre de pratiques complexes et désordonnées. Las preguntas acerca de cómo afectan los algoritmos a la (in)seguridad son cada vez más comunes en la Sociología Política Internacional. A fin de construir y sumar a estos debates, nuestro Debate Colectivo pone en primer plano las preguntas sobre la violencia algorítmica. Sostenemos que es importante abrir el debate acerca de cómo los sistemas algorítmicos alimentan (en) formas específicas de violencia, cómo justifican las acciones violentas o redefinen qué formas de violencia se consideran legítimas. A partir de la reunión de diferentes puntos de vista disciplinarios y conceptuales, este Debate Colectivo abre una conversación sobre la violencia algorítmica centrándose tanto en sus instancias específicas como en los desafíos de su conceptualización y estudio. En general, el debate converge en tres áreas de interés: la violencia que sustenta la creación y alimentación de las infraestructuras de datos, los procesos de traducción en juego en la utilización de la visión de la computadora/máquina a través de diversas prácticas de seguridad y el gobierno institucional de la violencia algorítmica, especialmente su organización, limitación y legitimación. Nuestro doble objetivo es mostrar el potencial de una conversación interdisciplinaria y avanzar hacia una agenda de investigación interactiva. Si bien nuestros abordajes divergen, se enriquecen mutuamente. Finalmente, destacamos la adquisición fundamental del estudio de las funciones de la violencia algorítmica en el tejido de lo internacional a través de un análisis situado de los sistemas algorítmicos como parte de prácticas complejas y, a menudo, desordenadas.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 }, { "id": "oa-https-doi-org-10-1080-01402390-2023-2241648", "title": "We’ll never have a model of an AI major-general: Artificial Intelligence, command decisions, and kitsch visions of war", "authors_or_org": "Cameron Hunter, Bleddyn E. Bowen", "year": 2023, "source_type": "paper", "bucket": "palantir", "url": "https://openalex.org/W4385663754", "doi_or_identifier": "10.1080/01402390.2023.2241648", "venue_or_site": "Journal of Strategic Studies", "abstract_or_summary": "Military AI optimists predict future AI assisting or making command decisions. We instead argue that, at a fundamental level, these predictions are dangerously wrong. The nature of war demands decisions based on abductive logic, whilst machine learning (or 'narrow AI') relies on inductive logic. The two forms of logic are not interchangeable, and therefore AI's limited utility in command - both tactical and strategic - is not something that can be solved by more data or more computing power. Many defence and government leaders are therefore proceeding with a false view of the nature of AI and of war itself.", "key_claims": [], "ontology_relevance": "", "ai_relevance": "", "palantir_relevance": "", "quality_signal": "scholarly_index", "retrieval_tags": [ "openalex", "palantir" ], "triage_tier": "candidate", "triage_score": 40 } ] }