{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WB7JHNSXCJMA3EHUYT75MTZBCD","short_pith_number":"pith:WB7JHNSX","schema_version":"1.0","canonical_sha256":"b07e93b65712580d90f4c4ffd64f2110f61c52fd0cec2b9f0cd41906bc2d2f87","source":{"kind":"arxiv","id":"2402.09267","version":2},"attestation_state":"computed","paper":{"title":"Self-Alignment for Factuality: Mitigating Hallucinations in LLMs via Self-Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baolin Peng, Haitao Mi, Helen Meng, Jingyan Zhou, Lifeng Jin, Linfeng Song, Xiaoying Zhang, Ye Tian","submitted_at":"2024-02-14T15:52:42Z","abstract_excerpt":"Despite showing increasingly human-like abilities, large language models (LLMs) often struggle with factual inaccuracies, i.e. \"hallucinations\", even when they hold relevant knowledge. To address these hallucinations, current approaches typically necessitate high-quality human factuality annotations. In this work, we explore Self-Alignment for Factuality, where we leverage the self-evaluation capability of an LLM to provide training signals that steer the model towards factuality. Specifically, we incorporate Self-Eval, a self-evaluation component, to prompt an LLM to validate the factuality o"},"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":"2402.09267","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-14T15:52:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9e4e0c233268e9e07612c541d2f91d2b60c8d2a4883d72a8e7ef445fa979061a","abstract_canon_sha256":"d21105897b363a06233d323acea3a8b5167562610013a057216c4604a78d7684"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:30:08.740360Z","signature_b64":"RYWt7dOe7dJ1FGtVct+5i4+mIz4jPmhcK/m9n0EdMGwveLTG2AB9lh0a48cp5j6tMQglPQ6LXE3S1E6nqnNBBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b07e93b65712580d90f4c4ffd64f2110f61c52fd0cec2b9f0cd41906bc2d2f87","last_reissued_at":"2026-07-05T08:30:08.739838Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:30:08.739838Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Alignment for Factuality: Mitigating Hallucinations in LLMs via Self-Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Baolin Peng, Haitao Mi, Helen Meng, Jingyan Zhou, Lifeng Jin, Linfeng Song, Xiaoying Zhang, Ye Tian","submitted_at":"2024-02-14T15:52:42Z","abstract_excerpt":"Despite showing increasingly human-like abilities, large language models (LLMs) often struggle with factual inaccuracies, i.e. \"hallucinations\", even when they hold relevant knowledge. To address these hallucinations, current approaches typically necessitate high-quality human factuality annotations. In this work, we explore Self-Alignment for Factuality, where we leverage the self-evaluation capability of an LLM to provide training signals that steer the model towards factuality. Specifically, we incorporate Self-Eval, a self-evaluation component, to prompt an LLM to validate the factuality o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09267","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/2402.09267/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":"2402.09267","created_at":"2026-07-05T08:30:08.739893+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.09267v2","created_at":"2026-07-05T08:30:08.739893+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09267","created_at":"2026-07-05T08:30:08.739893+00:00"},{"alias_kind":"pith_short_12","alias_value":"WB7JHNSXCJMA","created_at":"2026-07-05T08:30:08.739893+00:00"},{"alias_kind":"pith_short_16","alias_value":"WB7JHNSXCJMA3EHU","created_at":"2026-07-05T08:30:08.739893+00:00"},{"alias_kind":"pith_short_8","alias_value":"WB7JHNSX","created_at":"2026-07-05T08:30:08.739893+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2409.14634","citing_title":"Human-LLM Compound System for Scientific Ideation through Facet Recombination and Novelty Evaluation","ref_index":79,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD","json":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD.json","graph_json":"https://pith.science/api/pith-number/WB7JHNSXCJMA3EHUYT75MTZBCD/graph.json","events_json":"https://pith.science/api/pith-number/WB7JHNSXCJMA3EHUYT75MTZBCD/events.json","paper":"https://pith.science/paper/WB7JHNSX"},"agent_actions":{"view_html":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD","download_json":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD.json","view_paper":"https://pith.science/paper/WB7JHNSX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.09267&json=true","fetch_graph":"https://pith.science/api/pith-number/WB7JHNSXCJMA3EHUYT75MTZBCD/graph.json","fetch_events":"https://pith.science/api/pith-number/WB7JHNSXCJMA3EHUYT75MTZBCD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD/action/storage_attestation","attest_author":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD/action/author_attestation","sign_citation":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD/action/citation_signature","submit_replication":"https://pith.science/pith/WB7JHNSXCJMA3EHUYT75MTZBCD/action/replication_record"}},"created_at":"2026-07-05T08:30:08.739893+00:00","updated_at":"2026-07-05T08:30:08.739893+00:00"}