{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NBCQPJX45JD4PPCQ7ZM7E47WDD","short_pith_number":"pith:NBCQPJX4","canonical_record":{"source":{"id":"2607.23458","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-26T04:43:53Z","cross_cats_sorted":[],"title_canon_sha256":"4c790f7341b095656d06ff555dc6942e001361d9d111353613cb5a091c02a2fd","abstract_canon_sha256":"491d808aefc5aa8ba09bd86bcfdd30d957d31342c490b3caf682ad2ad7bbf06b"},"schema_version":"1.0"},"canonical_sha256":"684507a6fcea47c7bc50fe59f273f618faac49b826a08170a5a3b4b60bfb8101","source":{"kind":"arxiv","id":"2607.23458","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.23458","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"arxiv_version","alias_value":"2607.23458v1","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23458","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_12","alias_value":"NBCQPJX45JD4","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_16","alias_value":"NBCQPJX45JD4PPCQ","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_8","alias_value":"NBCQPJX4","created_at":"2026-07-28T01:22:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NBCQPJX45JD4PPCQ7ZM7E47WDD","target":"record","payload":{"canonical_record":{"source":{"id":"2607.23458","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-26T04:43:53Z","cross_cats_sorted":[],"title_canon_sha256":"4c790f7341b095656d06ff555dc6942e001361d9d111353613cb5a091c02a2fd","abstract_canon_sha256":"491d808aefc5aa8ba09bd86bcfdd30d957d31342c490b3caf682ad2ad7bbf06b"},"schema_version":"1.0"},"canonical_sha256":"684507a6fcea47c7bc50fe59f273f618faac49b826a08170a5a3b4b60bfb8101","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:22:53.653200Z","signature_b64":"8of6IDksM1cEPNGQRcZYJSSbMoX+rCMcVJzmlmCAVez333A0cCWBewv2Fg42GTpQ9zVyPxy3uygUdZQRqobyBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"684507a6fcea47c7bc50fe59f273f618faac49b826a08170a5a3b4b60bfb8101","last_reissued_at":"2026-07-28T01:22:53.652360Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:22:53.652360Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.23458","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-28T01:22:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dTnbt/CgdloT0FCIfEi2XNcabb1635NEskvX/+e/nSM5oXuEDPxnEKKc8ZaQ9vV4rI8wo2/wYiTdAKbghS+rBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:37:38.872606Z"},"content_sha256":"2dfeae9598e2c89570bad85e3618c973b01c2190837404ee85a145649a7a31a1","schema_version":"1.0","event_id":"sha256:2dfeae9598e2c89570bad85e3618c973b01c2190837404ee85a145649a7a31a1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NBCQPJX45JD4PPCQ7ZM7E47WDD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Two Regimes of Chain-of-Thought Unfaithfulness: Behavioral Detection Fails Where Models Are Wrong","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dikshant Aryal, Nick Rahimi, Suramya R. Angdembay","submitted_at":"2026-07-26T04:43:53Z","abstract_excerpt":"Chain-of-thought (CoT) explanations support oversight only if they are faithful: the stated reasoning must actually produce the answer. Auditing black-box (behavioral) detection of unfaithful CoT against FaithCoT-Bench's human annotations, we find answer correctness structures the problem at every level. Answer incorrectness alone (an oracle diagnostic, not a deployable detector) outperforms every purpose-built signal (AUROC 0.696), because 69% of annotated unfaithfulness occurs on incorrect answers. Stratifying by correctness splits detection into two regimes: on correct answers, behavioral s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23458","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/2607.23458/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-28T01:22:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gp6bszDt9cTch8S2qovzocFyAU27R4mzAEp0WhBZX0RlsvMQ3kjY5wIJmYd3EJIBv5DSvW5AsG/8O35HqXt8Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T12:37:38.873034Z"},"content_sha256":"853a4ea0f7bace3ba367b4b93f8e05d1585d9b1bc0eb49f07c4b17891da8c5d6","schema_version":"1.0","event_id":"sha256:853a4ea0f7bace3ba367b4b93f8e05d1585d9b1bc0eb49f07c4b17891da8c5d6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/bundle.json","state_url":"https://pith.science/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T12:37:38Z","links":{"resolver":"https://pith.science/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD","bundle":"https://pith.science/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/bundle.json","state":"https://pith.science/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NBCQPJX45JD4PPCQ7ZM7E47WDD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NBCQPJX45JD4PPCQ7ZM7E47WDD","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":"491d808aefc5aa8ba09bd86bcfdd30d957d31342c490b3caf682ad2ad7bbf06b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-26T04:43:53Z","title_canon_sha256":"4c790f7341b095656d06ff555dc6942e001361d9d111353613cb5a091c02a2fd"},"schema_version":"1.0","source":{"id":"2607.23458","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.23458","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"arxiv_version","alias_value":"2607.23458v1","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23458","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_12","alias_value":"NBCQPJX45JD4","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_16","alias_value":"NBCQPJX45JD4PPCQ","created_at":"2026-07-28T01:22:53Z"},{"alias_kind":"pith_short_8","alias_value":"NBCQPJX4","created_at":"2026-07-28T01:22:53Z"}],"graph_snapshots":[{"event_id":"sha256:853a4ea0f7bace3ba367b4b93f8e05d1585d9b1bc0eb49f07c4b17891da8c5d6","target":"graph","created_at":"2026-07-28T01:22:53Z","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/2607.23458/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Chain-of-thought (CoT) explanations support oversight only if they are faithful: the stated reasoning must actually produce the answer. Auditing black-box (behavioral) detection of unfaithful CoT against FaithCoT-Bench's human annotations, we find answer correctness structures the problem at every level. Answer incorrectness alone (an oracle diagnostic, not a deployable detector) outperforms every purpose-built signal (AUROC 0.696), because 69% of annotated unfaithfulness occurs on incorrect answers. Stratifying by correctness splits detection into two regimes: on correct answers, behavioral s","authors_text":"Dikshant Aryal, Nick Rahimi, Suramya R. Angdembay","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-26T04:43:53Z","title":"Two Regimes of Chain-of-Thought Unfaithfulness: Behavioral Detection Fails Where Models Are Wrong"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23458","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":"sha256:2dfeae9598e2c89570bad85e3618c973b01c2190837404ee85a145649a7a31a1","target":"record","created_at":"2026-07-28T01:22:53Z","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":"491d808aefc5aa8ba09bd86bcfdd30d957d31342c490b3caf682ad2ad7bbf06b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-26T04:43:53Z","title_canon_sha256":"4c790f7341b095656d06ff555dc6942e001361d9d111353613cb5a091c02a2fd"},"schema_version":"1.0","source":{"id":"2607.23458","kind":"arxiv","version":1}},"canonical_sha256":"684507a6fcea47c7bc50fe59f273f618faac49b826a08170a5a3b4b60bfb8101","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"684507a6fcea47c7bc50fe59f273f618faac49b826a08170a5a3b4b60bfb8101","first_computed_at":"2026-07-28T01:22:53.652360Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-28T01:22:53.652360Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8of6IDksM1cEPNGQRcZYJSSbMoX+rCMcVJzmlmCAVez333A0cCWBewv2Fg42GTpQ9zVyPxy3uygUdZQRqobyBA==","signature_status":"signed_v1","signed_at":"2026-07-28T01:22:53.653200Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.23458","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2dfeae9598e2c89570bad85e3618c973b01c2190837404ee85a145649a7a31a1","sha256:853a4ea0f7bace3ba367b4b93f8e05d1585d9b1bc0eb49f07c4b17891da8c5d6"],"state_sha256":"57b6592f70a3b44173c2cc4bd23bc5d73415f2515ba40a901777a5e0bdc88d78"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xiydQzBRTE8e+USufjKK9PIXPxuZN4tJ3ZSkVjfOdIubEhfREx48Y//p8uMnJGnXzH5RtB4dqUNb7C/q9CFXCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T12:37:38.875605Z","bundle_sha256":"1af62d3b44b5fa24dcdcb3bb19bd7f1bdfb42014d4f35ac6dd9290be92b820c7"}}