{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AYTLSQJ66G7DC5JTCZIMPKSXJ5","short_pith_number":"pith:AYTLSQJ6","canonical_record":{"source":{"id":"2503.21821","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-26T06:21:56Z","cross_cats_sorted":[],"title_canon_sha256":"7de518dc8b95f80a224b41c6e3aaae72b95a07bc7c57cb701ad37a1213fd1040","abstract_canon_sha256":"c529773490e5c9969d3785d4d48d23a146fcb31e38f86b7dbb06351844523f17"},"schema_version":"1.0"},"canonical_sha256":"0626b9413ef1be3175331650c7aa574f7620a6456bc7277b85ef2cb459a5f734","source":{"kind":"arxiv","id":"2503.21821","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21821","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21821v1","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21821","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_12","alias_value":"AYTLSQJ66G7D","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_16","alias_value":"AYTLSQJ66G7DC5JT","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_8","alias_value":"AYTLSQJ6","created_at":"2026-05-22T01:03:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AYTLSQJ66G7DC5JTCZIMPKSXJ5","target":"record","payload":{"canonical_record":{"source":{"id":"2503.21821","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-26T06:21:56Z","cross_cats_sorted":[],"title_canon_sha256":"7de518dc8b95f80a224b41c6e3aaae72b95a07bc7c57cb701ad37a1213fd1040","abstract_canon_sha256":"c529773490e5c9969d3785d4d48d23a146fcb31e38f86b7dbb06351844523f17"},"schema_version":"1.0"},"canonical_sha256":"0626b9413ef1be3175331650c7aa574f7620a6456bc7277b85ef2cb459a5f734","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:42.356140Z","signature_b64":"RbGoaWmpOI66gO0zFoiB7Hv51b0tvzxk9nrtBLf492IDfUrPN0mKPaSGWBil7npSCeZoo4Vp8+3B1WGI0afYCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0626b9413ef1be3175331650c7aa574f7620a6456bc7277b85ef2cb459a5f734","last_reissued_at":"2026-05-22T01:03:42.355303Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:42.355303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.21821","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:03:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/G5hegi0slYT0S6nHewQ/tSGhZLP3i1EE7BKT5kz/g+xHUfp1UOGv4naPo2QOHhb4RUI/qZdWuo1hXbp15dlBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T12:38:22.572205Z"},"content_sha256":"ba433a68a712126e0e4f1e6ed501db436a06c89c30b71265d7e858fbfaf148c6","schema_version":"1.0","event_id":"sha256:ba433a68a712126e0e4f1e6ed501db436a06c89c30b71265d7e858fbfaf148c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AYTLSQJ66G7DC5JTCZIMPKSXJ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PHYSICS: Benchmarking Foundation Models on University-Level Physics Problem Solving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Arman Cohan, Chen Zhao, John Sous, Kaiyue Feng, Tianyu Yang, Yilun Zhao, Yixin Liu","submitted_at":"2025-03-26T06:21:56Z","abstract_excerpt":"We introduce PHYSICS, a comprehensive benchmark for university-level physics problem solving. It contains 1297 expert-annotated problems covering six core areas: classical mechanics, quantum mechanics, thermodynamics and statistical mechanics, electromagnetism, atomic physics, and optics. Each problem requires advanced physics knowledge and mathematical reasoning. We develop a robust automated evaluation system for precise and reliable validation. Our evaluation of leading foundation models reveals substantial limitations. Even the most advanced model, o3-mini, achieves only 59.9% accuracy, hi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21821","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/2503.21821/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:03:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m8LCD/OeAOOMe7CqMlsYskDzDv7JKqq/jsW24z/p6sqE1kISiHlkfBs7v2NuhX6HlYjmskUi4/upofPNzm/QDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-24T12:38:22.572946Z"},"content_sha256":"7d21e6d920d47be8bfc02d443b871bcae3541b749daccacee55c66d505762b07","schema_version":"1.0","event_id":"sha256:7d21e6d920d47be8bfc02d443b871bcae3541b749daccacee55c66d505762b07"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/bundle.json","state_url":"https://pith.science/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-24T12:38:22Z","links":{"resolver":"https://pith.science/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5","bundle":"https://pith.science/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/bundle.json","state":"https://pith.science/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AYTLSQJ66G7DC5JTCZIMPKSXJ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AYTLSQJ66G7DC5JTCZIMPKSXJ5","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":"c529773490e5c9969d3785d4d48d23a146fcb31e38f86b7dbb06351844523f17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-26T06:21:56Z","title_canon_sha256":"7de518dc8b95f80a224b41c6e3aaae72b95a07bc7c57cb701ad37a1213fd1040"},"schema_version":"1.0","source":{"id":"2503.21821","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21821","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21821v1","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21821","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_12","alias_value":"AYTLSQJ66G7D","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_16","alias_value":"AYTLSQJ66G7DC5JT","created_at":"2026-05-22T01:03:42Z"},{"alias_kind":"pith_short_8","alias_value":"AYTLSQJ6","created_at":"2026-05-22T01:03:42Z"}],"graph_snapshots":[{"event_id":"sha256:7d21e6d920d47be8bfc02d443b871bcae3541b749daccacee55c66d505762b07","target":"graph","created_at":"2026-05-22T01:03:42Z","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/2503.21821/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce PHYSICS, a comprehensive benchmark for university-level physics problem solving. It contains 1297 expert-annotated problems covering six core areas: classical mechanics, quantum mechanics, thermodynamics and statistical mechanics, electromagnetism, atomic physics, and optics. Each problem requires advanced physics knowledge and mathematical reasoning. We develop a robust automated evaluation system for precise and reliable validation. Our evaluation of leading foundation models reveals substantial limitations. Even the most advanced model, o3-mini, achieves only 59.9% accuracy, hi","authors_text":"Arman Cohan, Chen Zhao, John Sous, Kaiyue Feng, Tianyu Yang, Yilun Zhao, Yixin Liu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-26T06:21:56Z","title":"PHYSICS: Benchmarking Foundation Models on University-Level Physics Problem Solving"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21821","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:ba433a68a712126e0e4f1e6ed501db436a06c89c30b71265d7e858fbfaf148c6","target":"record","created_at":"2026-05-22T01:03:42Z","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":"c529773490e5c9969d3785d4d48d23a146fcb31e38f86b7dbb06351844523f17","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-03-26T06:21:56Z","title_canon_sha256":"7de518dc8b95f80a224b41c6e3aaae72b95a07bc7c57cb701ad37a1213fd1040"},"schema_version":"1.0","source":{"id":"2503.21821","kind":"arxiv","version":1}},"canonical_sha256":"0626b9413ef1be3175331650c7aa574f7620a6456bc7277b85ef2cb459a5f734","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0626b9413ef1be3175331650c7aa574f7620a6456bc7277b85ef2cb459a5f734","first_computed_at":"2026-05-22T01:03:42.355303Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:03:42.355303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RbGoaWmpOI66gO0zFoiB7Hv51b0tvzxk9nrtBLf492IDfUrPN0mKPaSGWBil7npSCeZoo4Vp8+3B1WGI0afYCg==","signature_status":"signed_v1","signed_at":"2026-05-22T01:03:42.356140Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.21821","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ba433a68a712126e0e4f1e6ed501db436a06c89c30b71265d7e858fbfaf148c6","sha256:7d21e6d920d47be8bfc02d443b871bcae3541b749daccacee55c66d505762b07"],"state_sha256":"0bfbc917098bee1bd9bb097259f4ace5dc714cbb73baba67df58da75444f85f0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B0B8ULhoTOUz0bcKUr7qUv4mLM91KgHx+jq3Al/jTrYbaD5ydBK6sbRtQ/ULyfljlYmpdAXTv5Ie0ZoWC8PGBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-24T12:38:22.576965Z","bundle_sha256":"57f28f869a7d746e557a18bb161ea514f29ce236cb1002d848be3ba8348fb6b7"}}