{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CZPCDEZNDLKDEQVN3UIZHS4D2T","short_pith_number":"pith:CZPCDEZN","schema_version":"1.0","canonical_sha256":"165e21932d1ad43242addd1193cb83d4fe96335ae3351cc54f65061b13e346ad","source":{"kind":"arxiv","id":"2606.03883","version":1},"attestation_state":"computed","paper":{"title":"Reasoning Structure of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Fabian Farestam, Fr\\'ed\\'eric Berdoz, Luca A. Lanzend\\\"orfer, Roger Wattenhofer","submitted_at":"2026-06-02T16:49:19Z","abstract_excerpt":"Large reasoning models (LRMs) are often evaluated using metrics such as final-answer accuracy or token count. However, identical scores on these metrics can hide fundamentally different reasoning structures. To address this limitation, we introduce a scalable LRM benchmark of logic puzzles and a pipeline that converts unstructured traces into verifiable reasoning graphs of claims and dependencies. This turns reasoning into a structured, measurable object whose topology can be quantitatively analyzed. Building on this, we define a reasoning efficiency metric that quantifies how concentrated the"},"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":"2606.03883","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T16:49:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"80856431902d920738721199ca2fbe616db642ee0e0c65a535d844eec045e501","abstract_canon_sha256":"522a9028eef2904b95afdad07a602ddb2b18f9ce19ac11ce4e3b1d0feab02d83"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:06:05.207975Z","signature_b64":"IkKYjriRHqpYTOmBQKjVF/r1o+MCGDc5iUX+sU5n4GdV3Orj7JA/RMUuThzTdnKB71EuNu/7hUng0PAwGf+/AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"165e21932d1ad43242addd1193cb83d4fe96335ae3351cc54f65061b13e346ad","last_reissued_at":"2026-06-03T02:06:05.207634Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:06:05.207634Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning Structure of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Fabian Farestam, Fr\\'ed\\'eric Berdoz, Luca A. Lanzend\\\"orfer, Roger Wattenhofer","submitted_at":"2026-06-02T16:49:19Z","abstract_excerpt":"Large reasoning models (LRMs) are often evaluated using metrics such as final-answer accuracy or token count. However, identical scores on these metrics can hide fundamentally different reasoning structures. To address this limitation, we introduce a scalable LRM benchmark of logic puzzles and a pipeline that converts unstructured traces into verifiable reasoning graphs of claims and dependencies. This turns reasoning into a structured, measurable object whose topology can be quantitatively analyzed. Building on this, we define a reasoning efficiency metric that quantifies how concentrated the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03883","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/2606.03883/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":"2606.03883","created_at":"2026-06-03T02:06:05.207690+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03883v1","created_at":"2026-06-03T02:06:05.207690+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03883","created_at":"2026-06-03T02:06:05.207690+00:00"},{"alias_kind":"pith_short_12","alias_value":"CZPCDEZNDLKD","created_at":"2026-06-03T02:06:05.207690+00:00"},{"alias_kind":"pith_short_16","alias_value":"CZPCDEZNDLKDEQVN","created_at":"2026-06-03T02:06:05.207690+00:00"},{"alias_kind":"pith_short_8","alias_value":"CZPCDEZN","created_at":"2026-06-03T02:06:05.207690+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T","json":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T.json","graph_json":"https://pith.science/api/pith-number/CZPCDEZNDLKDEQVN3UIZHS4D2T/graph.json","events_json":"https://pith.science/api/pith-number/CZPCDEZNDLKDEQVN3UIZHS4D2T/events.json","paper":"https://pith.science/paper/CZPCDEZN"},"agent_actions":{"view_html":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T","download_json":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T.json","view_paper":"https://pith.science/paper/CZPCDEZN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03883&json=true","fetch_graph":"https://pith.science/api/pith-number/CZPCDEZNDLKDEQVN3UIZHS4D2T/graph.json","fetch_events":"https://pith.science/api/pith-number/CZPCDEZNDLKDEQVN3UIZHS4D2T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T/action/storage_attestation","attest_author":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T/action/author_attestation","sign_citation":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T/action/citation_signature","submit_replication":"https://pith.science/pith/CZPCDEZNDLKDEQVN3UIZHS4D2T/action/replication_record"}},"created_at":"2026-06-03T02:06:05.207690+00:00","updated_at":"2026-06-03T02:06:05.207690+00:00"}