{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NTDL4KZD5I3W4M74XRDIT7MEFL","short_pith_number":"pith:NTDL4KZD","schema_version":"1.0","canonical_sha256":"6cc6be2b23ea376e33fcbc4689fd842afba0a015ed20ef926efe4ab4163cebfa","source":{"kind":"arxiv","id":"2211.08412","version":2},"attestation_state":"computed","paper":{"title":"Evaluating the Factual Consistency of Large Language Models Through News Summarization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anisha Mascarenhas, Colin Raffel, Derek Tam, Mohit Bansal, Sarah Kwan, Shiyue Zhang","submitted_at":"2022-11-15T18:50:34Z","abstract_excerpt":"While large language models (LLMs) have proven to be effective on a large variety of tasks, they are also known to hallucinate information. To measure whether an LLM prefers factually consistent continuations of its input, we propose a new benchmark called FIB(Factual Inconsistency Benchmark) that focuses on the task of summarization. Specifically, our benchmark involves comparing the scores an LLM assigns to a factually consistent versus a factually inconsistent summary for an input news article. For factually consistent summaries, we use human-written reference summaries that we manually ver"},"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":"2211.08412","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-15T18:50:34Z","cross_cats_sorted":[],"title_canon_sha256":"3e372c623e1fce5ecb209b0f5867188a832fecd666880abd9c38043185befe07","abstract_canon_sha256":"3630d87f08a4b68ee8d4fa23eb08213a297c564fe50d2dfd825e1f4054dd4727"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:19:17.756398Z","signature_b64":"Bd4yn9xC/6YZUqsTzlgPCMptbWITh8Z6haldk918qvcEpFqtvBG654SbFWF4p++kvUpfgaLAzEWW3r5+IG3sCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6cc6be2b23ea376e33fcbc4689fd842afba0a015ed20ef926efe4ab4163cebfa","last_reissued_at":"2026-07-05T07:19:17.755901Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:19:17.755901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating the Factual Consistency of Large Language Models Through News Summarization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Anisha Mascarenhas, Colin Raffel, Derek Tam, Mohit Bansal, Sarah Kwan, Shiyue Zhang","submitted_at":"2022-11-15T18:50:34Z","abstract_excerpt":"While large language models (LLMs) have proven to be effective on a large variety of tasks, they are also known to hallucinate information. To measure whether an LLM prefers factually consistent continuations of its input, we propose a new benchmark called FIB(Factual Inconsistency Benchmark) that focuses on the task of summarization. Specifically, our benchmark involves comparing the scores an LLM assigns to a factually consistent versus a factually inconsistent summary for an input news article. For factually consistent summaries, we use human-written reference summaries that we manually ver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.08412","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/2211.08412/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":"2211.08412","created_at":"2026-07-05T07:19:17.755963+00:00"},{"alias_kind":"arxiv_version","alias_value":"2211.08412v2","created_at":"2026-07-05T07:19:17.755963+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.08412","created_at":"2026-07-05T07:19:17.755963+00:00"},{"alias_kind":"pith_short_12","alias_value":"NTDL4KZD5I3W","created_at":"2026-07-05T07:19:17.755963+00:00"},{"alias_kind":"pith_short_16","alias_value":"NTDL4KZD5I3W4M74","created_at":"2026-07-05T07:19:17.755963+00:00"},{"alias_kind":"pith_short_8","alias_value":"NTDL4KZD","created_at":"2026-07-05T07:19:17.755963+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.05741","citing_title":"Enhancing Factual Accuracy and Citation Generation in LLMs via Multi-Stage Self-Verification","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL","json":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL.json","graph_json":"https://pith.science/api/pith-number/NTDL4KZD5I3W4M74XRDIT7MEFL/graph.json","events_json":"https://pith.science/api/pith-number/NTDL4KZD5I3W4M74XRDIT7MEFL/events.json","paper":"https://pith.science/paper/NTDL4KZD"},"agent_actions":{"view_html":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL","download_json":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL.json","view_paper":"https://pith.science/paper/NTDL4KZD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2211.08412&json=true","fetch_graph":"https://pith.science/api/pith-number/NTDL4KZD5I3W4M74XRDIT7MEFL/graph.json","fetch_events":"https://pith.science/api/pith-number/NTDL4KZD5I3W4M74XRDIT7MEFL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL/action/storage_attestation","attest_author":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL/action/author_attestation","sign_citation":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL/action/citation_signature","submit_replication":"https://pith.science/pith/NTDL4KZD5I3W4M74XRDIT7MEFL/action/replication_record"}},"created_at":"2026-07-05T07:19:17.755963+00:00","updated_at":"2026-07-05T07:19:17.755963+00:00"}