{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3GJ36RWJM5Q75VPQAKOWQ7J3SP","short_pith_number":"pith:3GJ36RWJ","schema_version":"1.0","canonical_sha256":"d993bf46c96761fed5f0029d687d3b93d50e01d09ed6e316b9d0d6488e845588","source":{"kind":"arxiv","id":"2506.06020","version":1},"attestation_state":"computed","paper":{"title":"When to Trust Context: Self-Reflective Debates for Context Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amin Saberi, Cheng Meixin, Fang Wu, Haokai Zhao, Shayan Talaei, Tinson Xu, Yejin Choi, Zeqi Zhou","submitted_at":"2025-06-06T12:09:34Z","abstract_excerpt":"Large language models frequently encounter conflicts between their parametric knowledge and contextual input, often resulting in factual inconsistencies or hallucinations. We propose Self-Reflective Debate for Contextual Reliability (SR-DCR), a lightweight framework that integrates token-level self-confidence with an asymmetric multi-agent debate to adjudicate such conflicts. A critic, deprived of context, challenges a defender who argues from the given passage; a judge model evaluates the debate and determines the context's reliability. The final answer is selected by combining the verdict wi"},"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":"2506.06020","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-06T12:09:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"fb74a2ef3507cdc8d7048f6f762acce24de23b0f0579181944c134f2933b15b4","abstract_canon_sha256":"f2861c8b3cf545af3aa5ad84789ac710c0762e663f1631f6859b44836f061fe1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:14.772937Z","signature_b64":"UrdAKIR3AiY9nZ7JihGej77+B0TMmKWoEwzp4n0dWPmGDi5kDK0zfliPY0tpGBEUneOUq+Jgwyhv9DxWRBVsBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d993bf46c96761fed5f0029d687d3b93d50e01d09ed6e316b9d0d6488e845588","last_reissued_at":"2026-07-05T11:17:14.772430Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:14.772430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When to Trust Context: Self-Reflective Debates for Context Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Amin Saberi, Cheng Meixin, Fang Wu, Haokai Zhao, Shayan Talaei, Tinson Xu, Yejin Choi, Zeqi Zhou","submitted_at":"2025-06-06T12:09:34Z","abstract_excerpt":"Large language models frequently encounter conflicts between their parametric knowledge and contextual input, often resulting in factual inconsistencies or hallucinations. We propose Self-Reflective Debate for Contextual Reliability (SR-DCR), a lightweight framework that integrates token-level self-confidence with an asymmetric multi-agent debate to adjudicate such conflicts. A critic, deprived of context, challenges a defender who argues from the given passage; a judge model evaluates the debate and determines the context's reliability. The final answer is selected by combining the verdict wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06020","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/2506.06020/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":"2506.06020","created_at":"2026-07-05T11:17:14.772497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06020v1","created_at":"2026-07-05T11:17:14.772497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06020","created_at":"2026-07-05T11:17:14.772497+00:00"},{"alias_kind":"pith_short_12","alias_value":"3GJ36RWJM5Q7","created_at":"2026-07-05T11:17:14.772497+00:00"},{"alias_kind":"pith_short_16","alias_value":"3GJ36RWJM5Q75VPQ","created_at":"2026-07-05T11:17:14.772497+00:00"},{"alias_kind":"pith_short_8","alias_value":"3GJ36RWJ","created_at":"2026-07-05T11:17:14.772497+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP","json":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP.json","graph_json":"https://pith.science/api/pith-number/3GJ36RWJM5Q75VPQAKOWQ7J3SP/graph.json","events_json":"https://pith.science/api/pith-number/3GJ36RWJM5Q75VPQAKOWQ7J3SP/events.json","paper":"https://pith.science/paper/3GJ36RWJ"},"agent_actions":{"view_html":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP","download_json":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP.json","view_paper":"https://pith.science/paper/3GJ36RWJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06020&json=true","fetch_graph":"https://pith.science/api/pith-number/3GJ36RWJM5Q75VPQAKOWQ7J3SP/graph.json","fetch_events":"https://pith.science/api/pith-number/3GJ36RWJM5Q75VPQAKOWQ7J3SP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP/action/storage_attestation","attest_author":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP/action/author_attestation","sign_citation":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP/action/citation_signature","submit_replication":"https://pith.science/pith/3GJ36RWJM5Q75VPQAKOWQ7J3SP/action/replication_record"}},"created_at":"2026-07-05T11:17:14.772497+00:00","updated_at":"2026-07-05T11:17:14.772497+00:00"}