{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:YPQBUAO3V3RFYFKOFFTBJTZMRA","short_pith_number":"pith:YPQBUAO3","schema_version":"1.0","canonical_sha256":"c3e01a01dbaee25c154e296614cf2c883bd80565f74a134adae43677f3a4afe8","source":{"kind":"arxiv","id":"2305.13281","version":1},"attestation_state":"computed","paper":{"title":"LM vs LM: Detecting Factual Errors via Cross Examination","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amir Globerson, May Hamri, Mor Geva, Roi Cohen","submitted_at":"2023-05-22T17:42:14Z","abstract_excerpt":"A prominent weakness of modern language models (LMs) is their tendency to generate factually incorrect text, which hinders their usability. A natural question is whether such factual errors can be detected automatically. Inspired by truth-seeking mechanisms in law, we propose a factuality evaluation framework for LMs that is based on cross-examination. Our key idea is that an incorrect claim is likely to result in inconsistency with other claims that the model generates. To discover such inconsistencies, we facilitate a multi-turn interaction between the LM that generated the claim and another"},"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":"2305.13281","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T17:42:14Z","cross_cats_sorted":[],"title_canon_sha256":"830b62660a7e634535fd6c2a01c0e9a1780c8a70177fe5fe0846719d72d22a91","abstract_canon_sha256":"256a32962876cc0b3a6563d42c8fbb2c98484862f71809773da439d0a9a4366b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:28.551923Z","signature_b64":"DO8bREm9jm+RXyXzvC90CGUY1W8S90ullDpCFDS6/A7oDgVPHktd3XNDKNRCOtLlLgze2GRG1q/HLJHb53I0Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3e01a01dbaee25c154e296614cf2c883bd80565f74a134adae43677f3a4afe8","last_reissued_at":"2026-07-05T06:12:28.551420Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:28.551420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LM vs LM: Detecting Factual Errors via Cross Examination","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Amir Globerson, May Hamri, Mor Geva, Roi Cohen","submitted_at":"2023-05-22T17:42:14Z","abstract_excerpt":"A prominent weakness of modern language models (LMs) is their tendency to generate factually incorrect text, which hinders their usability. A natural question is whether such factual errors can be detected automatically. Inspired by truth-seeking mechanisms in law, we propose a factuality evaluation framework for LMs that is based on cross-examination. Our key idea is that an incorrect claim is likely to result in inconsistency with other claims that the model generates. To discover such inconsistencies, we facilitate a multi-turn interaction between the LM that generated the claim and another"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13281","kind":"arxiv","version":1},"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/2305.13281/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":"2305.13281","created_at":"2026-07-05T06:12:28.551481+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.13281v1","created_at":"2026-07-05T06:12:28.551481+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13281","created_at":"2026-07-05T06:12:28.551481+00:00"},{"alias_kind":"pith_short_12","alias_value":"YPQBUAO3V3RF","created_at":"2026-07-05T06:12:28.551481+00:00"},{"alias_kind":"pith_short_16","alias_value":"YPQBUAO3V3RFYFKO","created_at":"2026-07-05T06:12:28.551481+00:00"},{"alias_kind":"pith_short_8","alias_value":"YPQBUAO3","created_at":"2026-07-05T06:12:28.551481+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19057","citing_title":"Quantifying and Auditing LLM Evaluation via Positive--Unlabeled Learning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11127","citing_title":"Provenance-Grounded Gating and Adaptive Recovery in Synthetic Post-Training Data Curation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2309.11495","citing_title":"Chain-of-Verification Reduces Hallucination in Large Language Models","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2511.22409","citing_title":"NOMAD: A Multi-Agent LLM System for UML Class Diagram Generation from Natural Language Requirements","ref_index":22,"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":64,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA","json":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA.json","graph_json":"https://pith.science/api/pith-number/YPQBUAO3V3RFYFKOFFTBJTZMRA/graph.json","events_json":"https://pith.science/api/pith-number/YPQBUAO3V3RFYFKOFFTBJTZMRA/events.json","paper":"https://pith.science/paper/YPQBUAO3"},"agent_actions":{"view_html":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA","download_json":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA.json","view_paper":"https://pith.science/paper/YPQBUAO3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.13281&json=true","fetch_graph":"https://pith.science/api/pith-number/YPQBUAO3V3RFYFKOFFTBJTZMRA/graph.json","fetch_events":"https://pith.science/api/pith-number/YPQBUAO3V3RFYFKOFFTBJTZMRA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA/action/storage_attestation","attest_author":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA/action/author_attestation","sign_citation":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA/action/citation_signature","submit_replication":"https://pith.science/pith/YPQBUAO3V3RFYFKOFFTBJTZMRA/action/replication_record"}},"created_at":"2026-07-05T06:12:28.551481+00:00","updated_at":"2026-07-05T06:12:28.551481+00:00"}