{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KWVM7BT4QNAG336BPKP5PMNP4C","short_pith_number":"pith:KWVM7BT4","canonical_record":{"source":{"id":"2504.06260","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-08T17:59:39Z","cross_cats_sorted":["cs.CL","cs.NA","math.NA"],"title_canon_sha256":"7774304ce5575cfe8cdeb7c1a26307c8d98ae03fe7222330c2caac1c857862c1","abstract_canon_sha256":"65078f5f86bb4303478d54b63513cc0552694d57fe2b9fa878c1570d3fbd2546"},"schema_version":"1.0"},"canonical_sha256":"55aacf867c83406defc17a9fd7b1afe0a5dd75422fad556ece0eb83082b68587","source":{"kind":"arxiv","id":"2504.06260","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.06260","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.06260v1","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06260","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_12","alias_value":"KWVM7BT4QNAG","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_16","alias_value":"KWVM7BT4QNAG336B","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_8","alias_value":"KWVM7BT4","created_at":"2026-07-05T10:46:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KWVM7BT4QNAG336BPKP5PMNP4C","target":"record","payload":{"canonical_record":{"source":{"id":"2504.06260","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-08T17:59:39Z","cross_cats_sorted":["cs.CL","cs.NA","math.NA"],"title_canon_sha256":"7774304ce5575cfe8cdeb7c1a26307c8d98ae03fe7222330c2caac1c857862c1","abstract_canon_sha256":"65078f5f86bb4303478d54b63513cc0552694d57fe2b9fa878c1570d3fbd2546"},"schema_version":"1.0"},"canonical_sha256":"55aacf867c83406defc17a9fd7b1afe0a5dd75422fad556ece0eb83082b68587","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:18.144897Z","signature_b64":"OxTh4HY5rSIhYT+KZ8gW646on8hLP4KGNY77aGP1eUI0F1AlbROvJWIWcDFLBDYMl3hgjW8PSc22oL/0aUkRDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55aacf867c83406defc17a9fd7b1afe0a5dd75422fad556ece0eb83082b68587","last_reissued_at":"2026-07-05T10:46:18.144395Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:18.144395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.06260","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":"2026-07-05T10:46:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LwhH0q3Br7GrEjKK6ws+MjXoAh1rcoA/wwtw8Ci1whRyPXABZoJ2Ogmq9pdbLMRhakCchZ2PYVCYyLlaCrQrDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T20:59:06.246229Z"},"content_sha256":"03d079b80eaa63378430bb1f19a03688c0452354f313ecc8b2a8a93789a34fa4","schema_version":"1.0","event_id":"sha256:03d079b80eaa63378430bb1f19a03688c0452354f313ecc8b2a8a93789a34fa4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KWVM7BT4QNAG336BPKP5PMNP4C","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"FEABench: Evaluating Language Models on Multiphysics Reasoning Ability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.NA","math.NA"],"primary_cat":"cs.AI","authors_text":"Hao Cui, Michael P. Brenner, Nayantara Mudur, Paul Raccuglia, Peter Norgaard, Subhashini Venugopalan","submitted_at":"2025-04-08T17:59:39Z","abstract_excerpt":"Building precise simulations of the real world and invoking numerical solvers to answer quantitative problems is an essential requirement in engineering and science. We present FEABench, a benchmark to evaluate the ability of large language models (LLMs) and LLM agents to simulate and solve physics, mathematics and engineering problems using finite element analysis (FEA). We introduce a comprehensive evaluation scheme to investigate the ability of LLMs to solve these problems end-to-end by reasoning over natural language problem descriptions and operating COMSOL Multiphysics$^\\circledR$, an FE"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06260","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/2504.06260/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":"2026-07-05T10:46:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BexN49A+M0BLfmzK6Q3GpPLAMSWiTxpwXMZnkayM0QhTp7c7KBKLRMctWheLX9pLx22LSWqLvUTxZHunubh5BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T20:59:06.246733Z"},"content_sha256":"a05c44ff89c236efc8cf60071434824dd2d4aa34aae88ce09570fcb1a54100ef","schema_version":"1.0","event_id":"sha256:a05c44ff89c236efc8cf60071434824dd2d4aa34aae88ce09570fcb1a54100ef"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KWVM7BT4QNAG336BPKP5PMNP4C/bundle.json","state_url":"https://pith.science/pith/KWVM7BT4QNAG336BPKP5PMNP4C/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KWVM7BT4QNAG336BPKP5PMNP4C/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":"2026-08-06T20:59:06Z","links":{"resolver":"https://pith.science/pith/KWVM7BT4QNAG336BPKP5PMNP4C","bundle":"https://pith.science/pith/KWVM7BT4QNAG336BPKP5PMNP4C/bundle.json","state":"https://pith.science/pith/KWVM7BT4QNAG336BPKP5PMNP4C/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KWVM7BT4QNAG336BPKP5PMNP4C/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KWVM7BT4QNAG336BPKP5PMNP4C","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":"65078f5f86bb4303478d54b63513cc0552694d57fe2b9fa878c1570d3fbd2546","cross_cats_sorted":["cs.CL","cs.NA","math.NA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-08T17:59:39Z","title_canon_sha256":"7774304ce5575cfe8cdeb7c1a26307c8d98ae03fe7222330c2caac1c857862c1"},"schema_version":"1.0","source":{"id":"2504.06260","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.06260","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.06260v1","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06260","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_12","alias_value":"KWVM7BT4QNAG","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_16","alias_value":"KWVM7BT4QNAG336B","created_at":"2026-07-05T10:46:18Z"},{"alias_kind":"pith_short_8","alias_value":"KWVM7BT4","created_at":"2026-07-05T10:46:18Z"}],"graph_snapshots":[{"event_id":"sha256:a05c44ff89c236efc8cf60071434824dd2d4aa34aae88ce09570fcb1a54100ef","target":"graph","created_at":"2026-07-05T10:46:18Z","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/2504.06260/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Building precise simulations of the real world and invoking numerical solvers to answer quantitative problems is an essential requirement in engineering and science. We present FEABench, a benchmark to evaluate the ability of large language models (LLMs) and LLM agents to simulate and solve physics, mathematics and engineering problems using finite element analysis (FEA). We introduce a comprehensive evaluation scheme to investigate the ability of LLMs to solve these problems end-to-end by reasoning over natural language problem descriptions and operating COMSOL Multiphysics$^\\circledR$, an FE","authors_text":"Hao Cui, Michael P. Brenner, Nayantara Mudur, Paul Raccuglia, Peter Norgaard, Subhashini Venugopalan","cross_cats":["cs.CL","cs.NA","math.NA"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-08T17:59:39Z","title":"FEABench: Evaluating Language Models on Multiphysics Reasoning Ability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06260","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":"sha256:03d079b80eaa63378430bb1f19a03688c0452354f313ecc8b2a8a93789a34fa4","target":"record","created_at":"2026-07-05T10:46:18Z","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":"65078f5f86bb4303478d54b63513cc0552694d57fe2b9fa878c1570d3fbd2546","cross_cats_sorted":["cs.CL","cs.NA","math.NA"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-08T17:59:39Z","title_canon_sha256":"7774304ce5575cfe8cdeb7c1a26307c8d98ae03fe7222330c2caac1c857862c1"},"schema_version":"1.0","source":{"id":"2504.06260","kind":"arxiv","version":1}},"canonical_sha256":"55aacf867c83406defc17a9fd7b1afe0a5dd75422fad556ece0eb83082b68587","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55aacf867c83406defc17a9fd7b1afe0a5dd75422fad556ece0eb83082b68587","first_computed_at":"2026-07-05T10:46:18.144395Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:46:18.144395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OxTh4HY5rSIhYT+KZ8gW646on8hLP4KGNY77aGP1eUI0F1AlbROvJWIWcDFLBDYMl3hgjW8PSc22oL/0aUkRDw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:46:18.144897Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.06260","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:03d079b80eaa63378430bb1f19a03688c0452354f313ecc8b2a8a93789a34fa4","sha256:a05c44ff89c236efc8cf60071434824dd2d4aa34aae88ce09570fcb1a54100ef"],"state_sha256":"2d277bd1063c3c89585dfcda55819dacc7e65c64a372b8ce5c1097b1ffd3d754"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0/m5j90t79WgP+UhCk9KhPFK3QepQxT/EhlvTyci32B8yZWFRkIby3Yax7wlKE7ryEQO/qz79wcufdzgzhEkAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T20:59:06.250099Z","bundle_sha256":"92e391e7ad24500b87cf75986237605f84921ec627c672323b1cf77518586159"}}