{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Y42PHZ7CFSNAONY5C5XHW5D56A","short_pith_number":"pith:Y42PHZ7C","schema_version":"1.0","canonical_sha256":"c734f3e7e22c9a07371d176e7b747df0277d44860df33126d79c8df4c05a14e2","source":{"kind":"arxiv","id":"2305.14251","version":2},"attestation_state":"computed","paper":{"title":"FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hannaneh Hajishirzi, Kalpesh Krishna, Luke Zettlemoyer, Mike Lewis, Mohit Iyyer, Pang Wei Koh, Sewon Min, Wen-tau Yih, Xinxi Lyu","submitted_at":"2023-05-23T17:06:00Z","abstract_excerpt":"Evaluating the factuality of long-form text generated by large language models (LMs) is non-trivial because (1) generations often contain a mixture of supported and unsupported pieces of information, making binary judgments of quality inadequate, and (2) human evaluation is time-consuming and costly. In this paper, we introduce FACTSCORE, a new evaluation that breaks a generation into a series of atomic facts and computes the percentage of atomic facts supported by a reliable knowledge source. We conduct an extensive human evaluation to obtain FACTSCOREs of people biographies generated by seve"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2305.14251","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T17:06:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"eed7f4ddec5dde80a7bcf1868ac7f576ed08aa7d75e4ec51747808e6e52f5b6f","abstract_canon_sha256":"c67897e1b5ca33d7dfac87001167a692d674067ce75c9bb66db724d1a700b459"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:59:32.206854Z","signature_b64":"UsYbezGwPWz2sKraiMTnODbygNCl2zlOWifb5X3Aymc5PHsufgrDoKd3MrTwjnftGKmlBOPHXlQ+rGJfsTDOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c734f3e7e22c9a07371d176e7b747df0277d44860df33126d79c8df4c05a14e2","last_reissued_at":"2026-07-05T06:59:32.206376Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:59:32.206376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hannaneh Hajishirzi, Kalpesh Krishna, Luke Zettlemoyer, Mike Lewis, Mohit Iyyer, Pang Wei Koh, Sewon Min, Wen-tau Yih, Xinxi Lyu","submitted_at":"2023-05-23T17:06:00Z","abstract_excerpt":"Evaluating the factuality of long-form text generated by large language models (LMs) is non-trivial because (1) generations often contain a mixture of supported and unsupported pieces of information, making binary judgments of quality inadequate, and (2) human evaluation is time-consuming and costly. In this paper, we introduce FACTSCORE, a new evaluation that breaks a generation into a series of atomic facts and computes the percentage of atomic facts supported by a reliable knowledge source. We conduct an extensive human evaluation to obtain FACTSCOREs of people biographies generated by seve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14251","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2305.14251/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2305.14251","created_at":"2026-07-05T06:59:32.206444+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14251v2","created_at":"2026-07-05T06:59:32.206444+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14251","created_at":"2026-07-05T06:59:32.206444+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y42PHZ7CFSNA","created_at":"2026-07-05T06:59:32.206444+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y42PHZ7CFSNAONY5","created_at":"2026-07-05T06:59:32.206444+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y42PHZ7C","created_at":"2026-07-05T06:59:32.206444+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":27,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2607.08268","citing_title":"Different Teachers, Different Capabilities: Sub-1B On-Device Distillation for Structured Text Enrichment","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2607.08700","citing_title":"Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2607.05391","citing_title":"LLM-as-a-Verifier: A General-Purpose Verification Framework","ref_index":91,"is_internal_anchor":true},{"citing_arxiv_id":"2606.17391","citing_title":"NarrativeWorldBench: A Frontier-Saturated Benchmark and a Latent World Model for Long-Horizon Co-Creative Audio Drama","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13397","citing_title":"Mod-Guide: An LLM-based Content Moderation Feedback System to Address Insensitive Speech toward Indigenous Ethnic and Religious Minority Communities","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18285","citing_title":"RELIANCE: Curating and Evaluating Reproductive Health Information on Social Media","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07936","citing_title":"Illusions of the Gold Standard: A Large-scale Analysis of Human Evaluation Protocols for Long-form Text Generation","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00605","citing_title":"Auditing Forgetting in Limited Memory Language Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04435","citing_title":"Cascading Hallucination in Agentic RAG: The CHARM Framework for Detection and Mitigation","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15416","citing_title":"Margin-Adaptive Confidence Ranking for Reliable LLM Judgement","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00609","citing_title":"CARE-RL: Capability-Aware Reinforcement Learning for Mitigating Cross-Domain Conflicts","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24894","citing_title":"RWGBench: Evaluating Scholarly Positioning in Related Work Generation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02443","citing_title":"HalluScan: A Systematic Benchmark for Detecting and Mitigating Hallucinations in Instruction-Following LLMs","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2512.24366","citing_title":"On the Factual Consistency of Text-based Explainable Recommendation Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17487","citing_title":"Answer Only as Precisely as Justified: Calibrated Claim-Level Specificity Control for Agentic Systems","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2509.25868","citing_title":"ReFACT: A Benchmark for Scientific Confabulation Detection with Positional Error Annotations","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2309.11495","citing_title":"Chain-of-Verification Reduces Hallucination in Large Language Models","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11161","citing_title":"Althea: Human-AI Collaboration for Fact-Checking and Critical Reasoning","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2303.08896","citing_title":"SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12813","citing_title":"REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations","ref_index":178,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04947","citing_title":"SUMMIR: A Hallucination-Aware Framework for Ranking Sports Insights from LLMs","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2311.05232","citing_title":"A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions","ref_index":225,"is_internal_anchor":false},{"citing_arxiv_id":"2309.01219","citing_title":"Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2310.11511","citing_title":"Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflection","ref_index":136,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17487","citing_title":"Answer Only as Precisely as Justified: Calibrated Claim-Level Specificity Control for Agentic Systems","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A","json":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A.json","graph_json":"https://pith.science/api/pith-number/Y42PHZ7CFSNAONY5C5XHW5D56A/graph.json","events_json":"https://pith.science/api/pith-number/Y42PHZ7CFSNAONY5C5XHW5D56A/events.json","paper":"https://pith.science/paper/Y42PHZ7C"},"agent_actions":{"view_html":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A","download_json":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A.json","view_paper":"https://pith.science/paper/Y42PHZ7C","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14251&json=true","fetch_graph":"https://pith.science/api/pith-number/Y42PHZ7CFSNAONY5C5XHW5D56A/graph.json","fetch_events":"https://pith.science/api/pith-number/Y42PHZ7CFSNAONY5C5XHW5D56A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A/action/storage_attestation","attest_author":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A/action/author_attestation","sign_citation":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A/action/citation_signature","submit_replication":"https://pith.science/pith/Y42PHZ7CFSNAONY5C5XHW5D56A/action/replication_record"}},"created_at":"2026-07-05T06:59:32.206444+00:00","updated_at":"2026-07-05T06:59:32.206444+00:00"}