{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2TFAUQGA5YMIG4MNNSQDAZKZ66","short_pith_number":"pith:2TFAUQGA","schema_version":"1.0","canonical_sha256":"d4ca0a40c0ee1883718d6ca0306559f7957123883b1a98ea52e264794802bc8c","source":{"kind":"arxiv","id":"2605.29087","version":1},"attestation_state":"computed","paper":{"title":"The Chain Holds, the Answer Folds: Trace-Answer Dissociation in Reasoning Models Under Adversarial Pressure","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ramayya Krishnan, Rema Padman, Yubo Li","submitted_at":"2026-05-27T20:41:08Z","abstract_excerpt":"Reasoning models are evaluated on single-turn benchmarks but deployed in multi-turn dialogue, where users push back on correct answers. Under sustained adversarial pressure we find a previously undocumented failure mode: the chain-of-thought stays factually correct from first turn to last while the emitted answer flips wrong. We call this unfaithful capitulation (UC) and isolate it with a $2\\times 2$ latent-versus-behavioral framework that flip-rate metrics and single-turn faithfulness probes both miss. Across three datasets (MT-Consistency, MMLU-Pro, GSM8K), the latent-correct rate at the beh"},"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":"2605.29087","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T20:41:08Z","cross_cats_sorted":[],"title_canon_sha256":"98433f394aaf119718da7b82f13cff9a4799b28fbb01616a1dd93451e596e8da","abstract_canon_sha256":"d994f9113b3e927d4de9904bc5c20772a6e53f24a6fd04135960c92ddd1d3369"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:17.601976Z","signature_b64":"X36EVGqiD+Lp1wS5LfjMiUN3jfh9M+wjwwvYqG0WUkrMd0qGKrj4SkR0DXnKZv0ru4VO7vFHiym9GOew6PADCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4ca0a40c0ee1883718d6ca0306559f7957123883b1a98ea52e264794802bc8c","last_reissued_at":"2026-05-29T01:05:17.600920Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:17.600920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Chain Holds, the Answer Folds: Trace-Answer Dissociation in Reasoning Models Under Adversarial Pressure","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ramayya Krishnan, Rema Padman, Yubo Li","submitted_at":"2026-05-27T20:41:08Z","abstract_excerpt":"Reasoning models are evaluated on single-turn benchmarks but deployed in multi-turn dialogue, where users push back on correct answers. Under sustained adversarial pressure we find a previously undocumented failure mode: the chain-of-thought stays factually correct from first turn to last while the emitted answer flips wrong. We call this unfaithful capitulation (UC) and isolate it with a $2\\times 2$ latent-versus-behavioral framework that flip-rate metrics and single-turn faithfulness probes both miss. Across three datasets (MT-Consistency, MMLU-Pro, GSM8K), the latent-correct rate at the beh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29087","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/2605.29087/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":"2605.29087","created_at":"2026-05-29T01:05:17.601005+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29087v1","created_at":"2026-05-29T01:05:17.601005+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29087","created_at":"2026-05-29T01:05:17.601005+00:00"},{"alias_kind":"pith_short_12","alias_value":"2TFAUQGA5YMI","created_at":"2026-05-29T01:05:17.601005+00:00"},{"alias_kind":"pith_short_16","alias_value":"2TFAUQGA5YMIG4MN","created_at":"2026-05-29T01:05:17.601005+00:00"},{"alias_kind":"pith_short_8","alias_value":"2TFAUQGA","created_at":"2026-05-29T01:05:17.601005+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/2TFAUQGA5YMIG4MNNSQDAZKZ66","json":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66.json","graph_json":"https://pith.science/api/pith-number/2TFAUQGA5YMIG4MNNSQDAZKZ66/graph.json","events_json":"https://pith.science/api/pith-number/2TFAUQGA5YMIG4MNNSQDAZKZ66/events.json","paper":"https://pith.science/paper/2TFAUQGA"},"agent_actions":{"view_html":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66","download_json":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66.json","view_paper":"https://pith.science/paper/2TFAUQGA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29087&json=true","fetch_graph":"https://pith.science/api/pith-number/2TFAUQGA5YMIG4MNNSQDAZKZ66/graph.json","fetch_events":"https://pith.science/api/pith-number/2TFAUQGA5YMIG4MNNSQDAZKZ66/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66/action/storage_attestation","attest_author":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66/action/author_attestation","sign_citation":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66/action/citation_signature","submit_replication":"https://pith.science/pith/2TFAUQGA5YMIG4MNNSQDAZKZ66/action/replication_record"}},"created_at":"2026-05-29T01:05:17.601005+00:00","updated_at":"2026-05-29T01:05:17.601005+00:00"}