{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:E57R7TF6VBRILNSTTYFBU4CPFW","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":"99e4cf276e8b3bcec3991fb514d20cd157ea38d1d88447601cbe4c9c98c591de","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-01T06:42:02Z","title_canon_sha256":"4875d868d3661a3df972f7b3a98d5495e8e571e29665f64dcdd6572a1a944190"},"schema_version":"1.0","source":{"id":"2502.00334","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.00334","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.00334v4","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00334","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_12","alias_value":"E57R7TF6VBRI","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_16","alias_value":"E57R7TF6VBRILNST","created_at":"2026-07-05T11:14:53Z"},{"alias_kind":"pith_short_8","alias_value":"E57R7TF6","created_at":"2026-07-05T11:14:53Z"}],"graph_snapshots":[{"event_id":"sha256:e31901c58fd4999da3dc7769c3da1136bd854fde8c5daa8b35f2476e9f61da49","target":"graph","created_at":"2026-07-05T11:14:53Z","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/2502.00334/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have demonstrated remarkable capabilities in solving complex reasoning tasks, particularly in mathematics. However, the domain of physics reasoning presents unique challenges that have received significantly less attention. Existing benchmarks often fall short in evaluating LLMs' abilities on the breadth and depth of undergraduate-level physics, underscoring the need for a comprehensive evaluation. To fill this gap, we introduce UGPhysics, a large-scale and comprehensive benchmark specifically designed to evaluate UnderGraduate-level Physics (UGPhysics) reasoning w","authors_text":"Can Yang, Jiaxin Zhang, Qiyun Xu, Shizhe Diao, Tianhao Chen, Tong Xiao, Xin Xu, Yang Wang, Yuchen Yan","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-01T06:42:02Z","title":"UGPhysics: A Comprehensive Benchmark for Undergraduate Physics Reasoning with Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00334","kind":"arxiv","version":4},"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:df515cc0f25a0db54ee8271cbaad8353f2ab4faceae7b22fb2b48dfdccf26db1","target":"record","created_at":"2026-07-05T11:14:53Z","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":"99e4cf276e8b3bcec3991fb514d20cd157ea38d1d88447601cbe4c9c98c591de","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-01T06:42:02Z","title_canon_sha256":"4875d868d3661a3df972f7b3a98d5495e8e571e29665f64dcdd6572a1a944190"},"schema_version":"1.0","source":{"id":"2502.00334","kind":"arxiv","version":4}},"canonical_sha256":"277f1fccbea86285b6539e0a1a704f2d9424a38ffa6c199dbe2ac28583aa3d77","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"277f1fccbea86285b6539e0a1a704f2d9424a38ffa6c199dbe2ac28583aa3d77","first_computed_at":"2026-07-05T11:14:53.270642Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:53.270642Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"umv+lbv7fyuDAplRgPlF9L/uHzJWtgeIWDSvFLlcvy/+FSvQco2rT7FzkpeEaU+SIUFxbH1X1JIB2SoWFN/SBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:53.271156Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.00334","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:df515cc0f25a0db54ee8271cbaad8353f2ab4faceae7b22fb2b48dfdccf26db1","sha256:e31901c58fd4999da3dc7769c3da1136bd854fde8c5daa8b35f2476e9f61da49"],"state_sha256":"511dbb9155ca92a1d9be7d2de5bfe796e89947debf4f72d2c985161a9cc9d6e4"}