{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ESCC4IKRILPYVWD2R6Z2LRJYVK","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":"552acb01ab157a07bd8b8fb1877fb90cc44ae21af2bcfc6866a3fb9efbadf22b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-17T17:24:14Z","title_canon_sha256":"5b941e87aa6b66da51b7eeda79df5e3e947b69a61a79be3b7e0995aa8f2118e7"},"schema_version":"1.0","source":{"id":"2502.12054","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.12054","created_at":"2026-07-05T11:09:45Z"},{"alias_kind":"arxiv_version","alias_value":"2502.12054v2","created_at":"2026-07-05T11:09:45Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.12054","created_at":"2026-07-05T11:09:45Z"},{"alias_kind":"pith_short_12","alias_value":"ESCC4IKRILPY","created_at":"2026-07-05T11:09:45Z"},{"alias_kind":"pith_short_16","alias_value":"ESCC4IKRILPYVWD2","created_at":"2026-07-05T11:09:45Z"},{"alias_kind":"pith_short_8","alias_value":"ESCC4IKR","created_at":"2026-07-05T11:09:45Z"}],"graph_snapshots":[{"event_id":"sha256:a00565fc22c46d6becd92c14307351c0dc2dd0fb92d05a97d37d1969f67ab03f","target":"graph","created_at":"2026-07-05T11:09:45Z","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/2502.12054/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models demonstrate remarkable capabilities across various domains, especially mathematics and logic reasoning. However, current evaluations overlook physics-based reasoning - a complex task requiring physics theorems and constraints. We present PhysReason, a 1,200-problem benchmark comprising knowledge-based (25%) and reasoning-based (75%) problems, where the latter are divided into three difficulty levels (easy, medium, hard). Notably, problems require an average of 8.1 solution steps, with hard requiring 15.6, reflecting the complexity of physics-based reasoning. We propose th","authors_text":"Basura Fernando, Chengyou Jia, Jiaxing Huang, Jun Liu, Lingling Zhang, Mike Zheng Shou, Xinyu Zhang, Yanrui Wu, Yuxuan Dong","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-17T17:24:14Z","title":"PhysReason: A Comprehensive Benchmark towards Physics-Based Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.12054","kind":"arxiv","version":2},"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:abf1333dd594670fb1eb1493faa56da4feafde3635cfd4d74bcc9703ad98fce6","target":"record","created_at":"2026-07-05T11:09:45Z","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":"552acb01ab157a07bd8b8fb1877fb90cc44ae21af2bcfc6866a3fb9efbadf22b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-17T17:24:14Z","title_canon_sha256":"5b941e87aa6b66da51b7eeda79df5e3e947b69a61a79be3b7e0995aa8f2118e7"},"schema_version":"1.0","source":{"id":"2502.12054","kind":"arxiv","version":2}},"canonical_sha256":"24842e215142df8ad87a8fb3a5c538aa92318991f617e9a6ecec39fea6c4291f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"24842e215142df8ad87a8fb3a5c538aa92318991f617e9a6ecec39fea6c4291f","first_computed_at":"2026-07-05T11:09:45.309570Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:09:45.309570Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ElBT6f3K7i3Rwyzp26bQRR0pAPdBDBbyl111c+TwYePl8eeRRYkmsWStm1cqBCFwVhJskVukTguz7nww4S2GAg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:09:45.310083Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.12054","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:abf1333dd594670fb1eb1493faa56da4feafde3635cfd4d74bcc9703ad98fce6","sha256:a00565fc22c46d6becd92c14307351c0dc2dd0fb92d05a97d37d1969f67ab03f"],"state_sha256":"0f1d5f6b1a2b129faaa4a5dd0856a1ca190431a98dd23f940ad209d12d132808"}