{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7XCHDNJ7ONAYEURW5GM5IXKX3J","short_pith_number":"pith:7XCHDNJ7","canonical_record":{"source":{"id":"2605.28301","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:55:24Z","cross_cats_sorted":[],"title_canon_sha256":"349ac5e884accf1c74cc477ce8432ffb4fa996d35db9a38348d640a5007d4bac","abstract_canon_sha256":"6974a9e955d8949c29a990d9ad52ae5b595de030971357baba7d6c6abc88831d"},"schema_version":"1.0"},"canonical_sha256":"fdc471b53f7341825236e999d45d57da7cd514dd2259e062142897cc216f0888","source":{"kind":"arxiv","id":"2605.28301","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28301","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28301v1","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28301","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_12","alias_value":"7XCHDNJ7ONAY","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_16","alias_value":"7XCHDNJ7ONAYEURW","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_8","alias_value":"7XCHDNJ7","created_at":"2026-05-28T01:05:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7XCHDNJ7ONAYEURW5GM5IXKX3J","target":"record","payload":{"canonical_record":{"source":{"id":"2605.28301","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:55:24Z","cross_cats_sorted":[],"title_canon_sha256":"349ac5e884accf1c74cc477ce8432ffb4fa996d35db9a38348d640a5007d4bac","abstract_canon_sha256":"6974a9e955d8949c29a990d9ad52ae5b595de030971357baba7d6c6abc88831d"},"schema_version":"1.0"},"canonical_sha256":"fdc471b53f7341825236e999d45d57da7cd514dd2259e062142897cc216f0888","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:05:05.713678Z","signature_b64":"Pr9hK5HPxUJKQ3gL1s1vS6xY2MAaz0AbwBY0Er5dPLmLVozooOiEGMuyJ5qMIZso+KYcOtE6UI1uecVyf4I0DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fdc471b53f7341825236e999d45d57da7cd514dd2259e062142897cc216f0888","last_reissued_at":"2026-05-28T01:05:05.713253Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:05:05.713253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.28301","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":"2026-05-28T01:05:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RC4Hj49OZGD02ReWMMiGkDZbUX4gjD0Xk5QfPvMWN4rhwbRF6zUzow1d1qtfZsCPVN8WF0GNqxpnUeMR3vFICQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T11:29:18.713769Z"},"content_sha256":"138ff64ae53f228f3cbc972de815f3cb343a8c2da0f08ac0b90810b2bd1341ac","schema_version":"1.0","event_id":"sha256:138ff64ae53f228f3cbc972de815f3cb343a8c2da0f08ac0b90810b2bd1341ac"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7XCHDNJ7ONAYEURW5GM5IXKX3J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Better Accuracies, Worse Reasoning: A Step-Level Audit of Medical Chain-of-Thought Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fei Teng, Honghan Wu, Jiacong Mi, Xuanqi Peng, Yunsoo Kim, Zhaoyang Jiang, Zhizhong Fu, Zicheng Li","submitted_at":"2026-05-27T10:55:24Z","abstract_excerpt":"Chain-of-thought (CoT) distillation trains a smaller model to imitate a teacher's reasoning trace, but it is typically evaluated by final-answer metrics including accuracy. We ask whether gains in answer quality are accompanied by improvements in the trace. In medical QA, where short answer options can leave a richer clinical justification under-specified, a Qwen3-8B student distilled from a DeepSeek-V3-family teacher improves on MedQA-USMLE answer metrics (SC@64 74.7% to 84.4%; expected calibration error (ECE) 0.096 to 0.034). Yet under a Kimi-K2.6 style-blind LLM-judge audit, its error rate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28301","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/2605.28301/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":"2026-05-28T01:05:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DMHvt7XX97OXhIT8CR46d2wRgjSgd2w7uGsFetV5RO7FaCQ2TT3TUfPUOKWANJa/s5zmnGnXxMEdktIRp2AoCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T11:29:18.714583Z"},"content_sha256":"5d7d54703db978e829364b2aa04a497470df4050401c2351dc2b043b06eea9dd","schema_version":"1.0","event_id":"sha256:5d7d54703db978e829364b2aa04a497470df4050401c2351dc2b043b06eea9dd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/bundle.json","state_url":"https://pith.science/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/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":"2026-06-10T11:29:18Z","links":{"resolver":"https://pith.science/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J","bundle":"https://pith.science/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/bundle.json","state":"https://pith.science/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7XCHDNJ7ONAYEURW5GM5IXKX3J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7XCHDNJ7ONAYEURW5GM5IXKX3J","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":"6974a9e955d8949c29a990d9ad52ae5b595de030971357baba7d6c6abc88831d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:55:24Z","title_canon_sha256":"349ac5e884accf1c74cc477ce8432ffb4fa996d35db9a38348d640a5007d4bac"},"schema_version":"1.0","source":{"id":"2605.28301","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28301","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28301v1","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28301","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_12","alias_value":"7XCHDNJ7ONAY","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_16","alias_value":"7XCHDNJ7ONAYEURW","created_at":"2026-05-28T01:05:05Z"},{"alias_kind":"pith_short_8","alias_value":"7XCHDNJ7","created_at":"2026-05-28T01:05:05Z"}],"graph_snapshots":[{"event_id":"sha256:5d7d54703db978e829364b2aa04a497470df4050401c2351dc2b043b06eea9dd","target":"graph","created_at":"2026-05-28T01:05:05Z","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/2605.28301/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Chain-of-thought (CoT) distillation trains a smaller model to imitate a teacher's reasoning trace, but it is typically evaluated by final-answer metrics including accuracy. We ask whether gains in answer quality are accompanied by improvements in the trace. In medical QA, where short answer options can leave a richer clinical justification under-specified, a Qwen3-8B student distilled from a DeepSeek-V3-family teacher improves on MedQA-USMLE answer metrics (SC@64 74.7% to 84.4%; expected calibration error (ECE) 0.096 to 0.034). Yet under a Kimi-K2.6 style-blind LLM-judge audit, its error rate ","authors_text":"Fei Teng, Honghan Wu, Jiacong Mi, Xuanqi Peng, Yunsoo Kim, Zhaoyang Jiang, Zhizhong Fu, Zicheng Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:55:24Z","title":"Better Accuracies, Worse Reasoning: A Step-Level Audit of Medical Chain-of-Thought Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28301","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":"sha256:138ff64ae53f228f3cbc972de815f3cb343a8c2da0f08ac0b90810b2bd1341ac","target":"record","created_at":"2026-05-28T01:05:05Z","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":"6974a9e955d8949c29a990d9ad52ae5b595de030971357baba7d6c6abc88831d","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T10:55:24Z","title_canon_sha256":"349ac5e884accf1c74cc477ce8432ffb4fa996d35db9a38348d640a5007d4bac"},"schema_version":"1.0","source":{"id":"2605.28301","kind":"arxiv","version":1}},"canonical_sha256":"fdc471b53f7341825236e999d45d57da7cd514dd2259e062142897cc216f0888","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fdc471b53f7341825236e999d45d57da7cd514dd2259e062142897cc216f0888","first_computed_at":"2026-05-28T01:05:05.713253Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:05:05.713253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Pr9hK5HPxUJKQ3gL1s1vS6xY2MAaz0AbwBY0Er5dPLmLVozooOiEGMuyJ5qMIZso+KYcOtE6UI1uecVyf4I0DQ==","signature_status":"signed_v1","signed_at":"2026-05-28T01:05:05.713678Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28301","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:138ff64ae53f228f3cbc972de815f3cb343a8c2da0f08ac0b90810b2bd1341ac","sha256:5d7d54703db978e829364b2aa04a497470df4050401c2351dc2b043b06eea9dd"],"state_sha256":"a1c497329aa585634f6f55325bd651d07c9fabd5bb790b78a91c1136a89a61e9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9OQWiUvMpMpSirJWKstoZRP47HK9vu1FIqL3NtDvX87TfOMeW8avt6lG2GSXiSpHZ/LAE7Gz5GAETwv51lwNCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T11:29:18.719386Z","bundle_sha256":"37a75f075120f2dca59c6528b87452c10e19613526612ada0ab9555ef4badb92"}}