{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FFPE7BTQ6AO3J2JE77XWMUFW26","short_pith_number":"pith:FFPE7BTQ","schema_version":"1.0","canonical_sha256":"295e4f8670f01db4e924ffef6650b6d78ddf5e24f67866fbe7704fdc905b0fac","source":{"kind":"arxiv","id":"2506.02126","version":1},"attestation_state":"computed","paper":{"title":"Knowledge or Reasoning? A Close Look at How LLMs Think Across Domains","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cihang Xie, Hang Yu, Haoqin Tu, James Zou, Juncheng Wu, Sheng Liu, Xiaoke Huang, Yuyin Zhou","submitted_at":"2025-06-02T18:01:00Z","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"},"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":"2506.02126","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T18:01:00Z","cross_cats_sorted":[],"title_canon_sha256":"daad144af8588899901d7029370962be5a4b096b4c6f30d823b8e3cb5594f4fd","abstract_canon_sha256":"d3e0c22d0c910aed01b4a35b03d50a567703ba742c216b392d38c17c7b39979f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:33.022476Z","signature_b64":"Z3VpdkM1TFeAI1OctjnLbR9Ih7rHrJIxmWobtNsj1SzDfDy9FzjF9/XnS5IoA2g9WiCV3E1FtVsRjK6LEPUYDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"295e4f8670f01db4e924ffef6650b6d78ddf5e24f67866fbe7704fdc905b0fac","last_reissued_at":"2026-07-05T11:14:33.021879Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:33.021879Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Knowledge or Reasoning? A Close Look at How LLMs Think Across Domains","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cihang Xie, Hang Yu, Haoqin Tu, James Zou, Juncheng Wu, Sheng Liu, Xiaoke Huang, Yuyin Zhou","submitted_at":"2025-06-02T18:01:00Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.02126","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/2506.02126/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":"2506.02126","created_at":"2026-07-05T11:14:33.021940+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.02126v1","created_at":"2026-07-05T11:14:33.021940+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.02126","created_at":"2026-07-05T11:14:33.021940+00:00"},{"alias_kind":"pith_short_12","alias_value":"FFPE7BTQ6AO3","created_at":"2026-07-05T11:14:33.021940+00:00"},{"alias_kind":"pith_short_16","alias_value":"FFPE7BTQ6AO3J2JE","created_at":"2026-07-05T11:14:33.021940+00:00"},{"alias_kind":"pith_short_8","alias_value":"FFPE7BTQ","created_at":"2026-07-05T11:14:33.021940+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18453","citing_title":"LLM Parameters for Math Across Languages: Shared or Separate?","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01104","citing_title":"CausalMix: Data Mixture as Causal Inference for Language Model Training","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20176","citing_title":"ClinSeekAgent: Automating Multimodal Evidence Seeking for Agentic Clinical Reasoning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01853","citing_title":"Spatiotemporal Hidden-State Dynamics as a Signature of Internal Reasoning in Large Language Models","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26","json":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26.json","graph_json":"https://pith.science/api/pith-number/FFPE7BTQ6AO3J2JE77XWMUFW26/graph.json","events_json":"https://pith.science/api/pith-number/FFPE7BTQ6AO3J2JE77XWMUFW26/events.json","paper":"https://pith.science/paper/FFPE7BTQ"},"agent_actions":{"view_html":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26","download_json":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26.json","view_paper":"https://pith.science/paper/FFPE7BTQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.02126&json=true","fetch_graph":"https://pith.science/api/pith-number/FFPE7BTQ6AO3J2JE77XWMUFW26/graph.json","fetch_events":"https://pith.science/api/pith-number/FFPE7BTQ6AO3J2JE77XWMUFW26/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26/action/storage_attestation","attest_author":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26/action/author_attestation","sign_citation":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26/action/citation_signature","submit_replication":"https://pith.science/pith/FFPE7BTQ6AO3J2JE77XWMUFW26/action/replication_record"}},"created_at":"2026-07-05T11:14:33.021940+00:00","updated_at":"2026-07-05T11:14:33.021940+00:00"}