{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TLAYVI4VIVJ6NVLCN56BD5C77H","short_pith_number":"pith:TLAYVI4V","canonical_record":{"source":{"id":"2502.17214","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T14:48:06Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"c3c4c0b004f1bddcce201aa7b9eedbcaa01786420795493b927165ed0825af84","abstract_canon_sha256":"865324af08f2151dc8ff2aeaad467808d1f88ee4e69769333ed1bca9739059ed"},"schema_version":"1.0"},"canonical_sha256":"9ac18aa3954553e6d5626f7c11f45ff9d2f507ebd5e6f765bc4375be05ed3ea0","source":{"kind":"arxiv","id":"2502.17214","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.17214","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.17214v2","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17214","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"TLAYVI4VIVJ6","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"TLAYVI4VIVJ6NVLC","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"TLAYVI4V","created_at":"2026-07-05T11:14:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TLAYVI4VIVJ6NVLCN56BD5C77H","target":"record","payload":{"canonical_record":{"source":{"id":"2502.17214","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T14:48:06Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"c3c4c0b004f1bddcce201aa7b9eedbcaa01786420795493b927165ed0825af84","abstract_canon_sha256":"865324af08f2151dc8ff2aeaad467808d1f88ee4e69769333ed1bca9739059ed"},"schema_version":"1.0"},"canonical_sha256":"9ac18aa3954553e6d5626f7c11f45ff9d2f507ebd5e6f765bc4375be05ed3ea0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:27.691944Z","signature_b64":"7X+fSE3ayML7hItecJXpdW5kmTL/NCzREQdZBUmeu/DmvKVsqTOpXUX+x2ICvHFUdvG5H/GXBx7yB4HCsu7SCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ac18aa3954553e6d5626f7c11f45ff9d2f507ebd5e6f765bc4375be05ed3ea0","last_reissued_at":"2026-07-05T11:14:27.691495Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:27.691495Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.17214","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wUl/KIBspTf5Q2Lc/fD6vTJr0D/n8OA5uw5fOlO7bNj2agjq91LlmFATpN+87enLlBhTGoHJMn18ZSX1m6xqCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:03:08.960989Z"},"content_sha256":"0899769a1757b35fd9a1415be2b40a4bc68b27bd0e991c8dc72efb2c3869b39a","schema_version":"1.0","event_id":"sha256:0899769a1757b35fd9a1415be2b40a4bc68b27bd0e991c8dc72efb2c3869b39a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TLAYVI4VIVJ6NVLCN56BD5C77H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CoT-UQ: Improving Response-wise Uncertainty Quantification in LLMs with Chain-of-Thought","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Boxuan Zhang, Ruqi Zhang","submitted_at":"2025-02-24T14:48:06Z","abstract_excerpt":"Large language models (LLMs) excel in many tasks but struggle to accurately quantify uncertainty in their generated responses. This limitation makes it challenging to detect misinformation and ensure reliable decision-making. Existing uncertainty quantification (UQ) methods for LLMs are primarily prompt-wise rather than response-wise, often requiring multiple response samples, which incurs high computational costs. Moreover, LLMs have been shown to be overconfident, particularly when using reasoning steps to derive their answers. In this work, we propose CoT-UQ, a response-wise UQ framework th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17214","kind":"arxiv","version":2},"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/2502.17214/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":"2026-07-05T11:14:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2ldEhASIMdgR/4ihv6aIbipz9DTf5kYliM1MNeGdzIaKgYfcMy7V9NSJYtkEZEuE+rFsMbQsGlzcVOd5W656Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T18:03:08.961570Z"},"content_sha256":"28d3da076ebace68347a264ade419c8646e929628e0ce0971186979e62e892fd","schema_version":"1.0","event_id":"sha256:28d3da076ebace68347a264ade419c8646e929628e0ce0971186979e62e892fd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/bundle.json","state_url":"https://pith.science/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/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":"2026-08-08T18:03:08Z","links":{"resolver":"https://pith.science/pith/TLAYVI4VIVJ6NVLCN56BD5C77H","bundle":"https://pith.science/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/bundle.json","state":"https://pith.science/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TLAYVI4VIVJ6NVLCN56BD5C77H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TLAYVI4VIVJ6NVLCN56BD5C77H","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":"865324af08f2151dc8ff2aeaad467808d1f88ee4e69769333ed1bca9739059ed","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T14:48:06Z","title_canon_sha256":"c3c4c0b004f1bddcce201aa7b9eedbcaa01786420795493b927165ed0825af84"},"schema_version":"1.0","source":{"id":"2502.17214","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.17214","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"arxiv_version","alias_value":"2502.17214v2","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17214","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_12","alias_value":"TLAYVI4VIVJ6","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_16","alias_value":"TLAYVI4VIVJ6NVLC","created_at":"2026-07-05T11:14:27Z"},{"alias_kind":"pith_short_8","alias_value":"TLAYVI4V","created_at":"2026-07-05T11:14:27Z"}],"graph_snapshots":[{"event_id":"sha256:28d3da076ebace68347a264ade419c8646e929628e0ce0971186979e62e892fd","target":"graph","created_at":"2026-07-05T11:14:27Z","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/2502.17214/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) excel in many tasks but struggle to accurately quantify uncertainty in their generated responses. This limitation makes it challenging to detect misinformation and ensure reliable decision-making. Existing uncertainty quantification (UQ) methods for LLMs are primarily prompt-wise rather than response-wise, often requiring multiple response samples, which incurs high computational costs. Moreover, LLMs have been shown to be overconfident, particularly when using reasoning steps to derive their answers. In this work, we propose CoT-UQ, a response-wise UQ framework th","authors_text":"Boxuan Zhang, Ruqi Zhang","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T14:48:06Z","title":"CoT-UQ: Improving Response-wise Uncertainty Quantification in LLMs with Chain-of-Thought"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17214","kind":"arxiv","version":2},"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":"sha256:0899769a1757b35fd9a1415be2b40a4bc68b27bd0e991c8dc72efb2c3869b39a","target":"record","created_at":"2026-07-05T11:14:27Z","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":"865324af08f2151dc8ff2aeaad467808d1f88ee4e69769333ed1bca9739059ed","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-02-24T14:48:06Z","title_canon_sha256":"c3c4c0b004f1bddcce201aa7b9eedbcaa01786420795493b927165ed0825af84"},"schema_version":"1.0","source":{"id":"2502.17214","kind":"arxiv","version":2}},"canonical_sha256":"9ac18aa3954553e6d5626f7c11f45ff9d2f507ebd5e6f765bc4375be05ed3ea0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9ac18aa3954553e6d5626f7c11f45ff9d2f507ebd5e6f765bc4375be05ed3ea0","first_computed_at":"2026-07-05T11:14:27.691495Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:27.691495Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7X+fSE3ayML7hItecJXpdW5kmTL/NCzREQdZBUmeu/DmvKVsqTOpXUX+x2ICvHFUdvG5H/GXBx7yB4HCsu7SCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:27.691944Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.17214","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0899769a1757b35fd9a1415be2b40a4bc68b27bd0e991c8dc72efb2c3869b39a","sha256:28d3da076ebace68347a264ade419c8646e929628e0ce0971186979e62e892fd"],"state_sha256":"71ad841ebdb0904fceead6df356d42295720e6bafb0e69c3f2639e4966c7dab3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kuKTGj66tk5ly4t0LJq4az+TGiWQwc/bSl7dFuaA025SMgdKu+pk1T03HQBEzJ779Png4qi8XO+Z9i1AAjdbAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T18:03:08.965501Z","bundle_sha256":"f4bc344d30b01e8136e40aa17bba38bfaaf2f8e9fb2f6091a82b78650f86c698"}}