{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "ai-oct-facts-benchmark-family", "title": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "summary": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges.", "boundary": "One evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.", "area": "LLM factuality and grounded generation", "topic": "Artificial intelligence", "year": 2024, "lane": "HUMAN", "result": "BENCHMARK", "contribution_kind": "EVALUATION_RESOURCE_DESIGN", "package_id": "ai-oct-facts-benchmark-family", "target": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "humans": [ "Alon Jacovi", "Andrew Wang", "Chris Alberti", "Connie Tao", "Jon Lipovetz", "Kate Olszewska", "Lukas Haas", "Michelle Liu", "Nate Keating", "Adam Bloniarz", "Carl Saroufim", "Corey Fry", "Dror Marcus", "Doron Kukliansky", "Gaurav Singh Tomar", "James Swirhun", "Jinwei Xing", "Lily Wang", "Madhu Gurumurthy", "Michael Aaron", "Moran Ambar", "Rachana Fellinger", "Rui Wang", "Zizhao Zhang", "Sasha Goldshtein", "Dipanjan Das", "Aileen Cheng", "Amir Globerson", "Ben Golan", "Charles Kwong", "Eyal Ben-David", "Yonatan Bitton", "Aijun Bai", "Anfal Siddiqui", "Aravindan Raghuveer", "Arturo Bajuelos Castillo", "Aviel Atias", "Chang Liu", "Daniel Balle", "Deepanway Ghosal", "Elena Gribovskaya", "Eran Ofek", "Honglei Zhuang", "Itay Laish", "Jan Ackermann", "Meg Risdal", "Megan Barnes", "Michael Fink", "Mohamed Amin", "Natan Potikha", "Nikita Gupta", "Nitzan Katz", "Noam Velan", "Ofir Roval", "Ori Ram", "Polina Zablotskaia", "Prathamesh Bang", "Priyanka Agrawal", "Rakesh Ghiya", "Sanjay Ganapathy", "Simon Baumgartner", "Sofia Erell", "Sushant Prakash", "Thibault Sellam", "Vikram Rao", "Xuanhui Wang", "Yaroslav Akulov", "Yulong Yang", "Zhen Yang", "Zhixin Lai", "Zhongru Wu", "Anca Dragan", "Avinatan Hassidim", "Fernando Pereira", "Slav Petrov", "Srinivasan Venkatachary", "Tulsee Doshi", "Yossi Matias" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "ai-oct-src-facts-benchmark-family", "title": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" }, { "id": "ai-oct-src-facts-suite", "title": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" }, { "id": "ai-oct-src-facts-official-launch", "title": "FACTS Grounding: A new benchmark for evaluating the factuality of large language models", "url": "https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/" } ], "points": "85000", "published_grade": { "id": "ai-oct-facts-benchmark-family", "entry_id": "ai-oct-facts-benchmark-family", "band": "B", "modifier": "+", "target": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "what_changed": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges.", "scope": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges. One evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Alon Jacovi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Andrew Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Chris Alberti", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Connie Tao", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jon Lipovetz", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Kate Olszewska", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Lukas Haas", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Michelle Liu", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Nate Keating", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Adam Bloniarz", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Carl Saroufim", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Corey Fry", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Dror Marcus", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Doron Kukliansky", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Gaurav Singh Tomar", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "James Swirhun", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jinwei Xing", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Lily Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Madhu Gurumurthy", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Michael Aaron", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Moran Ambar", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Rachana Fellinger", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Rui Wang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Zizhao Zhang", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sasha Goldshtein", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Dipanjan Das", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Aileen Cheng", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Amir Globerson", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Ben Golan", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Charles Kwong", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Eyal Ben-David", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-1", "kind": "source_reference", "source": { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-2", "kind": "source_reference", "source": { "citation": "FACTS Grounding: A new benchmark for evaluating the factuality of large language models", "url": "https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "One evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.", "domain": "LLM factuality and grounded generation" }, "sources": [ { "citation": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1" }, { "citation": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1" }, { "citation": "FACTS Grounding: A new benchmark for evaluating the factuality of large language models", "url": "https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/" } ], "statement": "# The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input\n\nDocument-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges.\n\n## Scientific scope\n\nOne evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nAlon Jacovi; Andrew Wang; Chris Alberti; Connie Tao; Jon Lipovetz; Kate Olszewska; Lukas Haas; Michelle Liu; Nate Keating; Adam Bloniarz; Carl Saroufim; Corey Fry; Dror Marcus; Doron Kukliansky; Gaurav Singh Tomar; James Swirhun; Jinwei Xing; Lily Wang; Madhu Gurumurthy; Michael Aaron; Moran Ambar; Rachana Fellinger; Rui Wang; Zizhao Zhang; Sasha Goldshtein; Dipanjan Das; Aileen Cheng; Amir Globerson; Ben Golan; Charles Kwong; Eyal Ben-David; Yonatan Bitton; Aijun Bai; Anfal Siddiqui; Aravindan Raghuveer; Arturo Bajuelos Castillo; Aviel Atias; Chang Liu; Daniel Balle; Deepanway Ghosal; Elena Gribovskaya; Eran Ofek; Honglei Zhuang; Itay Laish; Jan Ackermann; Meg Risdal; Megan Barnes; Michael Fink; Mohamed Amin; Natan Potikha; Nikita Gupta; Nitzan Katz; Noam Velan; Ofir Roval; Ori Ram; Polina Zablotskaia; Prathamesh Bang; Priyanka Agrawal; Rakesh Ghiya; Sanjay Ganapathy; Simon Baumgartner; Sofia Erell; Sushant Prakash; Thibault Sellam; Vikram Rao; Xuanhui Wang; Yaroslav Akulov; Yulong Yang; Zhen Yang; Zhixin Lai; Zhongru Wu; Anca Dragan; Avinatan Hassidim; Fernando Pereira; Slav Petrov; Srinivasan Venkatachary; Tulsee Doshi; Yossi Matias\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input](https://arxiv.org/pdf/2501.03200v1)\n- [The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality](https://arxiv.org/pdf/2512.10791v1)\n- [FACTS Grounding: A new benchmark for evaluating the factuality of large language models](https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/)\n", "summary": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges.", "title": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input" }, "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "source_ref": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:604433ab89b08d4961a3539c7739909dc7283bb9470f8307b775394cd3784df3" }, "target": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "ai-oct-facts-benchmark-family", "entry_id": "ai-oct-facts-benchmark-family", "revision": 1, "title": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "context": "LLM factuality and grounded generation", "track": "Evaluation resource design", "band": "B", "modifier": "+", "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges.", "scope": "Document-grounding benchmark filters unhelpful responses before multi-judge support assessment; the later FACTS suite adds multimodal, parametric and standardized-search dimensions plus revised grounding judges. One evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. One evolving evaluation family with source-specific dates/bylines, not four additional inventions. Source grounding is not all-world truth, and benchmarks do not certify safety or universal reliability. Search access and judge protocols matter; original and revised grounding scores are not interchangeable.", "references": [ { "title": "The FACTS Grounding Leaderboard: Benchmarking LLMs' Ability to Ground Responses to Long-Form Input", "url": "https://arxiv.org/pdf/2501.03200v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." }, { "title": "The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality", "url": "https://arxiv.org/pdf/2512.10791v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." }, { "title": "FACTS Grounding: A new benchmark for evaluating the factuality of large language models", "url": "https://deepmind.google/discover/blog/facts-grounding-a-new-benchmark-for-evaluating-the-factuality-of-large-language-models/", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:70a139b9-2899-4faf-aede-17bfbd6829f2", "version": 1, "content_hash": "sha256:45945a4afb7f41c3e14d83e82f4fc9fece3561a5c1a213c07dd8bf422646f709" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "85000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }