{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IUBJNPEIW5ECZKER73KLSVFB4G","short_pith_number":"pith:IUBJNPEI","schema_version":"1.0","canonical_sha256":"450296bc88b7482ca891fed4b954a1e198670538a9872ed83014b9f0300a06c8","source":{"kind":"arxiv","id":"2607.24604","version":1},"attestation_state":"computed","paper":{"title":"Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jianwei Yang, Qiang Yang, Xueping Gao","submitted_at":"2026-07-27T16:05:23Z","abstract_excerpt":"Generate--test--revise loops are common in coding agents, but repetition alone provides no reliability guarantee. We study the gap between finding a correct patch and retaining, verifying, and submitting it. A sealed five-seed study over 30 HumanEval repairs produces 900 three-revision trajectories. Under forced revision, current correctness with current traces falls from 0.820 after one revision to 0.673 after two, although ever-correct rises to 0.847. Two common-state studies use 2,430 branches from identical frozen programs to remove post-treatment risk-set bias. In a prespecified 14B repli"},"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":"2607.24604","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-27T16:05:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"918128732853394ffb540042bcce1527079fb12ddafb4f702bdc5e8d476068c1","abstract_canon_sha256":"ec890f23a09f1651e132e726f12e329979c3534ec7000d087faa09002a3558c8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:24:12.553827Z","signature_b64":"krGz2Tr7B5R3yxqIyWKxBwW7ciyzcfYUlOuOs1TMSOEtYwlRV4uN4bfOjldQpsfGeYsvGwe2W+b7LLTqPBhjDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"450296bc88b7482ca891fed4b954a1e198670538a9872ed83014b9f0300a06c8","last_reissued_at":"2026-07-28T02:24:12.552993Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:24:12.552993Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jianwei Yang, Qiang Yang, Xueping Gao","submitted_at":"2026-07-27T16:05:23Z","abstract_excerpt":"Generate--test--revise loops are common in coding agents, but repetition alone provides no reliability guarantee. We study the gap between finding a correct patch and retaining, verifying, and submitting it. A sealed five-seed study over 30 HumanEval repairs produces 900 three-revision trajectories. Under forced revision, current correctness with current traces falls from 0.820 after one revision to 0.673 after two, although ever-correct rises to 0.847. Two common-state studies use 2,430 branches from identical frozen programs to remove post-treatment risk-set bias. In a prespecified 14B repli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24604","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/2607.24604/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":"2607.24604","created_at":"2026-07-28T02:24:12.553426+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.24604v1","created_at":"2026-07-28T02:24:12.553426+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24604","created_at":"2026-07-28T02:24:12.553426+00:00"},{"alias_kind":"pith_short_12","alias_value":"IUBJNPEIW5EC","created_at":"2026-07-28T02:24:12.553426+00:00"},{"alias_kind":"pith_short_16","alias_value":"IUBJNPEIW5ECZKER","created_at":"2026-07-28T02:24:12.553426+00:00"},{"alias_kind":"pith_short_8","alias_value":"IUBJNPEI","created_at":"2026-07-28T02:24:12.553426+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G","json":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G.json","graph_json":"https://pith.science/api/pith-number/IUBJNPEIW5ECZKER73KLSVFB4G/graph.json","events_json":"https://pith.science/api/pith-number/IUBJNPEIW5ECZKER73KLSVFB4G/events.json","paper":"https://pith.science/paper/IUBJNPEI"},"agent_actions":{"view_html":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G","download_json":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G.json","view_paper":"https://pith.science/paper/IUBJNPEI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.24604&json=true","fetch_graph":"https://pith.science/api/pith-number/IUBJNPEIW5ECZKER73KLSVFB4G/graph.json","fetch_events":"https://pith.science/api/pith-number/IUBJNPEIW5ECZKER73KLSVFB4G/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G/action/storage_attestation","attest_author":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G/action/author_attestation","sign_citation":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G/action/citation_signature","submit_replication":"https://pith.science/pith/IUBJNPEIW5ECZKER73KLSVFB4G/action/replication_record"}},"created_at":"2026-07-28T02:24:12.553426+00:00","updated_at":"2026-07-28T02:24:12.553426+00:00"}