{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FFPE7BTQ6AO3J2JE77XWMUFW26","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":"d3e0c22d0c910aed01b4a35b03d50a567703ba742c216b392d38c17c7b39979f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T18:01:00Z","title_canon_sha256":"daad144af8588899901d7029370962be5a4b096b4c6f30d823b8e3cb5594f4fd"},"schema_version":"1.0","source":{"id":"2506.02126","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.02126","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"arxiv_version","alias_value":"2506.02126v1","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02126","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_12","alias_value":"FFPE7BTQ6AO3","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_16","alias_value":"FFPE7BTQ6AO3J2JE","created_at":"2026-07-05T11:14:33Z"},{"alias_kind":"pith_short_8","alias_value":"FFPE7BTQ","created_at":"2026-07-05T11:14:33Z"}],"graph_snapshots":[{"event_id":"sha256:1767eae119d380d987d1d199c88524f55b287896a184f0dd49b82e82b4412d17","target":"graph","created_at":"2026-07-05T11:14:33Z","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/2506.02126/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in reasoning-enhanced Large Language Models such as OpenAI-o1/3 and DeepSeek-R1 have significantly improved performance on complex tasks. However, the quality and transparency of their internal reasoning processes remain underexplored. This work moves beyond the final-answer accuracy and investigates step-by-step reasoning in the medical and mathematical domains by explicitly decomposing the thinking trajectories into two parts: knowledge and reasoning. Specifically, we introduce a fine-grained evaluation framework that judges: (1) the correctness of knowledge used (measured by","authors_text":"Cihang Xie, Hang Yu, Haoqin Tu, James Zou, Juncheng Wu, Sheng Liu, Xiaoke Huang, Yuyin Zhou","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T18:01:00Z","title":"Knowledge or Reasoning? A Close Look at How LLMs Think Across Domains"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02126","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:a8d4d596b9e10bdbf025b5a9f6fc70c62a14b0834fac09fb2db4a7fb5e46ec50","target":"record","created_at":"2026-07-05T11:14:33Z","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":"d3e0c22d0c910aed01b4a35b03d50a567703ba742c216b392d38c17c7b39979f","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T18:01:00Z","title_canon_sha256":"daad144af8588899901d7029370962be5a4b096b4c6f30d823b8e3cb5594f4fd"},"schema_version":"1.0","source":{"id":"2506.02126","kind":"arxiv","version":1}},"canonical_sha256":"295e4f8670f01db4e924ffef6650b6d78ddf5e24f67866fbe7704fdc905b0fac","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"295e4f8670f01db4e924ffef6650b6d78ddf5e24f67866fbe7704fdc905b0fac","first_computed_at":"2026-07-05T11:14:33.021879Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:33.021879Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Z3VpdkM1TFeAI1OctjnLbR9Ih7rHrJIxmWobtNsj1SzDfDy9FzjF9/XnS5IoA2g9WiCV3E1FtVsRjK6LEPUYDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:33.022476Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.02126","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8d4d596b9e10bdbf025b5a9f6fc70c62a14b0834fac09fb2db4a7fb5e46ec50","sha256:1767eae119d380d987d1d199c88524f55b287896a184f0dd49b82e82b4412d17"],"state_sha256":"c6cc6bc7e74d42a065483f218ea544d8796d3d972e31863e192fa21de2ec71bc"}