{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RCCKXJI2BJUNNAYZCT752EME2Z","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":"052291f2daa675a5629625e3ef6b74c0d173e1b3073a09746246cd0af55f8357","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-12-02T18:11:30Z","title_canon_sha256":"e616f578cf5e62d8aaeca70e4d44365a52b66e0b2d19bc8b4a2bcee6a583dd2a"},"schema_version":"1.0","source":{"id":"2412.01769","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01769","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01769v1","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01769","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_12","alias_value":"RCCKXJI2BJUN","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_16","alias_value":"RCCKXJI2BJUNNAYZ","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_8","alias_value":"RCCKXJI2","created_at":"2026-07-05T09:43:19Z"}],"graph_snapshots":[{"event_id":"sha256:da33a5f49ba95c8d5c4f0a8b7daf90f7506f27d837f51b090d97f1714943badd","target":"graph","created_at":"2026-07-05T09:43:19Z","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/2412.01769/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the goal of benchmarking generative systems beyond expert software development ability, we introduce Commit0, a benchmark that challenges AI agents to write libraries from scratch. Agents are provided with a specification document outlining the library's API as well as a suite of interactive unit tests, with the goal of producing an implementation of this API accordingly. The implementation is validated through running these unit tests. As a benchmark, Commit0 is designed to move beyond static one-shot code generation towards agents that must process long-form natural language specificati","authors_text":"Alexander M Rush, Celine Lee, Claire Cardie, Justin T Chiu, Matthias Gall\\'e, Nan Jiang, Wenting Zhao","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-12-02T18:11:30Z","title":"Commit0: Library Generation from Scratch"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01769","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":"sha256:7cf5684bfc05f0e2cbf25c86487646a7bd41714268c4066ae476412c6701ccfb","target":"record","created_at":"2026-07-05T09:43:19Z","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":"052291f2daa675a5629625e3ef6b74c0d173e1b3073a09746246cd0af55f8357","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-12-02T18:11:30Z","title_canon_sha256":"e616f578cf5e62d8aaeca70e4d44365a52b66e0b2d19bc8b4a2bcee6a583dd2a"},"schema_version":"1.0","source":{"id":"2412.01769","kind":"arxiv","version":1}},"canonical_sha256":"8884aba51a0a68d6831914ffdd1184d6469e39a09360a2e842e5b6aacfbcd391","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8884aba51a0a68d6831914ffdd1184d6469e39a09360a2e842e5b6aacfbcd391","first_computed_at":"2026-07-05T09:43:19.227583Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:43:19.227583Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"S8uiWOhBUdh2o1z+9/bgZkN3S9O4H0P9oyDiUmKiaY5Y78uaTYbY4N+x7OwR2ypToJK0stFr+NPfdAwC0qTOAw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:43:19.228200Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.01769","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7cf5684bfc05f0e2cbf25c86487646a7bd41714268c4066ae476412c6701ccfb","sha256:da33a5f49ba95c8d5c4f0a8b7daf90f7506f27d837f51b090d97f1714943badd"],"state_sha256":"a35a041e3dd2ac605f9261766968bf9635b821b4aeb1be2a45bd1559c007330c"}