{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ONN7JVFKXZVL6BL345COSDOWQM","short_pith_number":"pith:ONN7JVFK","schema_version":"1.0","canonical_sha256":"735bf4d4aabe6abf057be744e90dd683108614d4fd465794659043ac6d92e754","source":{"kind":"arxiv","id":"2303.04910","version":2},"attestation_state":"computed","paper":{"title":"Baldur: Whole-Proof Generation and Repair with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LO","cs.SE"],"primary_cat":"cs.LG","authors_text":"Emily First, Markus N. Rabe, Talia Ringer, Yuriy Brun","submitted_at":"2023-03-08T22:00:15Z","abstract_excerpt":"Formally verifying software properties is a highly desirable but labor-intensive task. Recent work has developed methods to automate formal verification using proof assistants, such as Coq and Isabelle/HOL, e.g., by training a model to predict one proof step at a time, and using that model to search through the space of possible proofs. This paper introduces a new method to automate formal verification: We use large language models, trained on natural language text and code and fine-tuned on proofs, to generate whole proofs for theorems at once, rather than one step at a time. We combine this "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2303.04910","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-03-08T22:00:15Z","cross_cats_sorted":["cs.LO","cs.SE"],"title_canon_sha256":"3e8086312f6e062d9d12d1ecd38b10ab226fb516f204cb05cd939b0556017196","abstract_canon_sha256":"9078a04a675539e38af775d77e239119ca0a3e77c4f239bb5730cbe46f7b14cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:51:45.440384Z","signature_b64":"jQ2vV6W6F6XxbQX7aIUuwbBCtI8Q4qoKSBSIJvuwg91ucwXxlLr4nkxT1byyv7LFIHpmIhhUcOsMOnWKYsLrAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"735bf4d4aabe6abf057be744e90dd683108614d4fd465794659043ac6d92e754","last_reissued_at":"2026-07-05T05:51:45.440007Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:51:45.440007Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Baldur: Whole-Proof Generation and Repair with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LO","cs.SE"],"primary_cat":"cs.LG","authors_text":"Emily First, Markus N. Rabe, Talia Ringer, Yuriy Brun","submitted_at":"2023-03-08T22:00:15Z","abstract_excerpt":"Formally verifying software properties is a highly desirable but labor-intensive task. Recent work has developed methods to automate formal verification using proof assistants, such as Coq and Isabelle/HOL, e.g., by training a model to predict one proof step at a time, and using that model to search through the space of possible proofs. This paper introduces a new method to automate formal verification: We use large language models, trained on natural language text and code and fine-tuned on proofs, to generate whole proofs for theorems at once, rather than one step at a time. We combine this "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.04910","kind":"arxiv","version":2},"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/2303.04910/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2303.04910","created_at":"2026-07-05T05:51:45.440063+00:00"},{"alias_kind":"arxiv_version","alias_value":"2303.04910v2","created_at":"2026-07-05T05:51:45.440063+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.04910","created_at":"2026-07-05T05:51:45.440063+00:00"},{"alias_kind":"pith_short_12","alias_value":"ONN7JVFKXZVL","created_at":"2026-07-05T05:51:45.440063+00:00"},{"alias_kind":"pith_short_16","alias_value":"ONN7JVFKXZVL6BL3","created_at":"2026-07-05T05:51:45.440063+00:00"},{"alias_kind":"pith_short_8","alias_value":"ONN7JVFK","created_at":"2026-07-05T05:51:45.440063+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07779","citing_title":"From Solvers to Research: Large Language Model-Driven Formal Mathematics at the Research Frontier","ref_index":70,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09450","citing_title":"TheoremBench: Evaluating LLMs on Theorem Proving in Formal Mathematics","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2310.10631","citing_title":"Llemma: An Open Language Model For Mathematics","ref_index":138,"is_internal_anchor":false},{"citing_arxiv_id":"2511.12253","citing_title":"The Search for Constrained Random Generators","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM","json":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM.json","graph_json":"https://pith.science/api/pith-number/ONN7JVFKXZVL6BL345COSDOWQM/graph.json","events_json":"https://pith.science/api/pith-number/ONN7JVFKXZVL6BL345COSDOWQM/events.json","paper":"https://pith.science/paper/ONN7JVFK"},"agent_actions":{"view_html":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM","download_json":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM.json","view_paper":"https://pith.science/paper/ONN7JVFK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2303.04910&json=true","fetch_graph":"https://pith.science/api/pith-number/ONN7JVFKXZVL6BL345COSDOWQM/graph.json","fetch_events":"https://pith.science/api/pith-number/ONN7JVFKXZVL6BL345COSDOWQM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM/action/storage_attestation","attest_author":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM/action/author_attestation","sign_citation":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM/action/citation_signature","submit_replication":"https://pith.science/pith/ONN7JVFKXZVL6BL345COSDOWQM/action/replication_record"}},"created_at":"2026-07-05T05:51:45.440063+00:00","updated_at":"2026-07-05T05:51:45.440063+00:00"}