{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YPQBUAO3V3RFYFKOFFTBJTZMRA","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"256a32962876cc0b3a6563d42c8fbb2c98484862f71809773da439d0a9a4366b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T17:42:14Z","title_canon_sha256":"830b62660a7e634535fd6c2a01c0e9a1780c8a70177fe5fe0846719d72d22a91"},"schema_version":"1.0","source":{"id":"2305.13281","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.13281","created_at":"2026-07-05T06:12:28Z"},{"alias_kind":"arxiv_version","alias_value":"2305.13281v1","created_at":"2026-07-05T06:12:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13281","created_at":"2026-07-05T06:12:28Z"},{"alias_kind":"pith_short_12","alias_value":"YPQBUAO3V3RF","created_at":"2026-07-05T06:12:28Z"},{"alias_kind":"pith_short_16","alias_value":"YPQBUAO3V3RFYFKO","created_at":"2026-07-05T06:12:28Z"},{"alias_kind":"pith_short_8","alias_value":"YPQBUAO3","created_at":"2026-07-05T06:12:28Z"}],"graph_snapshots":[{"event_id":"sha256:f8a9e32e131034496aa3be65cc00688fc6e775a3779d64937dab33883c96e9bd","target":"graph","created_at":"2026-07-05T06:12:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2305.13281/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Amir Globerson, May Hamri, Mor Geva, Roi Cohen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T17:42:14Z","title":"LM vs LM: Detecting Factual Errors via Cross Examination"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13281","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:40143fc359c37ecdb5e84257ff22592cab069fa25b8ac3e633fafb50ba87b21c","target":"record","created_at":"2026-07-05T06:12:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"256a32962876cc0b3a6563d42c8fbb2c98484862f71809773da439d0a9a4366b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-22T17:42:14Z","title_canon_sha256":"830b62660a7e634535fd6c2a01c0e9a1780c8a70177fe5fe0846719d72d22a91"},"schema_version":"1.0","source":{"id":"2305.13281","kind":"arxiv","version":1}},"canonical_sha256":"c3e01a01dbaee25c154e296614cf2c883bd80565f74a134adae43677f3a4afe8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c3e01a01dbaee25c154e296614cf2c883bd80565f74a134adae43677f3a4afe8","first_computed_at":"2026-07-05T06:12:28.551420Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:12:28.551420Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"DO8bREm9jm+RXyXzvC90CGUY1W8S90ullDpCFDS6/A7oDgVPHktd3XNDKNRCOtLlLgze2GRG1q/HLJHb53I0Bw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:12:28.551923Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.13281","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:40143fc359c37ecdb5e84257ff22592cab069fa25b8ac3e633fafb50ba87b21c","sha256:f8a9e32e131034496aa3be65cc00688fc6e775a3779d64937dab33883c96e9bd"],"state_sha256":"0716d82fabd039296067de3ddcfe6b9f57c1e73c92c4c4bf3ab56c222693c0ab"}