{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7AIMDSR3YQHUQXRIODV3DMSSDN","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":"df3cbfe0b9a51cfe39088dc3171c204c1d0f318a86a36c58f4af56bd22ef98ad","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-04T22:00:52Z","title_canon_sha256":"8bee0379d9e0076b96cd90828f029e95a62f3744fdcafef37a8d147b9c4968dd"},"schema_version":"1.0","source":{"id":"2506.04481","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04481","created_at":"2026-07-05T11:16:15Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04481v1","created_at":"2026-07-05T11:16:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04481","created_at":"2026-07-05T11:16:15Z"},{"alias_kind":"pith_short_12","alias_value":"7AIMDSR3YQHU","created_at":"2026-07-05T11:16:15Z"},{"alias_kind":"pith_short_16","alias_value":"7AIMDSR3YQHUQXRI","created_at":"2026-07-05T11:16:15Z"},{"alias_kind":"pith_short_8","alias_value":"7AIMDSR3","created_at":"2026-07-05T11:16:15Z"}],"graph_snapshots":[{"event_id":"sha256:971176e8eea01988621a147c2c73d9dc5a1c2925be2529211c24ad741b28082e","target":"graph","created_at":"2026-07-05T11:16:15Z","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/2506.04481/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Although large language models (LLMs) show promise in solving complex mathematical tasks, existing evaluation paradigms rely solely on a coarse measure of overall answer accuracy, which are insufficient for assessing their authentic capabilities. In this paper, we propose \\textbf{CogMath}, which comprehensively assesses LLMs' mathematical abilities through the lens of human cognition. Specifically, inspired by psychological theories, CogMath formalizes human reasoning process into 3 stages: \\emph{problem comprehension}, \\emph{problem solving}, and \\emph{solution summarization}. Within these st","authors_text":"Cheng cheng, Enhong Chen, Jiayu Liu, Jing Sha, Jinze Wu, Qi Liu, Shijin Wang, Song Li, Wei Dai, Zhenya Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-04T22:00:52Z","title":"CogMath: Assessing LLMs' Authentic Mathematical Ability from a Human Cognitive Perspective"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04481","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":"sha256:7a9f6278f105349c66fbc3162831697fbc70d1faa460d58871b5c706bb0afba5","target":"record","created_at":"2026-07-05T11:16:15Z","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":"df3cbfe0b9a51cfe39088dc3171c204c1d0f318a86a36c58f4af56bd22ef98ad","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-06-04T22:00:52Z","title_canon_sha256":"8bee0379d9e0076b96cd90828f029e95a62f3744fdcafef37a8d147b9c4968dd"},"schema_version":"1.0","source":{"id":"2506.04481","kind":"arxiv","version":1}},"canonical_sha256":"f810c1ca3bc40f485e2870ebb1b2521b719d7e1aa7ce504be418f70254807925","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f810c1ca3bc40f485e2870ebb1b2521b719d7e1aa7ce504be418f70254807925","first_computed_at":"2026-07-05T11:16:15.294569Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:16:15.294569Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y1uYo7aZLMRpGwMRfcJ+WK0FYIuiPBBDmYIyXvHqc3djNkcSPkvlSBTgEhICRmQUZQrPbA3yBP8dkLR7XP+RBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:16:15.295054Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04481","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7a9f6278f105349c66fbc3162831697fbc70d1faa460d58871b5c706bb0afba5","sha256:971176e8eea01988621a147c2c73d9dc5a1c2925be2529211c24ad741b28082e"],"state_sha256":"27933d3bb19255bdd84131f24d141e24b92d575852ff03ff4d853e0706ef8a3f"}