{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7ZQDKRBOIDKM4QEDZIQ5FHZCTH","short_pith_number":"pith:7ZQDKRBO","schema_version":"1.0","canonical_sha256":"fe6035442e40d4ce4083ca21d29f2299f6e39800d39073744ca56209ec516948","source":{"kind":"arxiv","id":"2506.17772","version":1},"attestation_state":"computed","paper":{"title":"PAGENT: Learning to Patch Software Engineering Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Gias Uddin, Haoran Xue, Song Wang","submitted_at":"2025-06-21T18:00:00Z","abstract_excerpt":"LLM Agents produce patches automatically to resolve an issue. However, they can generate inaccurate patches. Little is known about the root causes behind those failed patches or how those could be fixed. This paper reports an empirical study of the failed patches generated by seven top LLM code agents. We collected 114 issues from the SWE-bench Lite dataset that remained unresolved across the agents. The seven agents produced a total of 769 failed patches for those issues, which we checked with a combination of GPT-4o and manual analysis. We present a taxonomy of the failure reasons across the"},"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":"2506.17772","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-06-21T18:00:00Z","cross_cats_sorted":[],"title_canon_sha256":"039e4eaf335c062c02297787505c5f07bc682d17b08dbc0efd95718bd041597a","abstract_canon_sha256":"4e8d1a7107f00bf7d9fa611e65f0982ce16c0f5e282ef13ac8e883f25c5af6c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:22.998805Z","signature_b64":"T4FIZ5I+gJ2eNXqdziyX2rtYHiBpJ9jLyAR+uRhDGm20biL93164YezSGhuRYjrgetcgSnYMojnljk3WXwzSAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe6035442e40d4ce4083ca21d29f2299f6e39800d39073744ca56209ec516948","last_reissued_at":"2026-07-05T11:25:22.998376Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:22.998376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PAGENT: Learning to Patch Software Engineering Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Gias Uddin, Haoran Xue, Song Wang","submitted_at":"2025-06-21T18:00:00Z","abstract_excerpt":"LLM Agents produce patches automatically to resolve an issue. However, they can generate inaccurate patches. Little is known about the root causes behind those failed patches or how those could be fixed. This paper reports an empirical study of the failed patches generated by seven top LLM code agents. We collected 114 issues from the SWE-bench Lite dataset that remained unresolved across the agents. The seven agents produced a total of 769 failed patches for those issues, which we checked with a combination of GPT-4o and manual analysis. We present a taxonomy of the failure reasons across the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17772","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/2506.17772/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":"2506.17772","created_at":"2026-07-05T11:25:22.998452+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.17772v1","created_at":"2026-07-05T11:25:22.998452+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17772","created_at":"2026-07-05T11:25:22.998452+00:00"},{"alias_kind":"pith_short_12","alias_value":"7ZQDKRBOIDKM","created_at":"2026-07-05T11:25:22.998452+00:00"},{"alias_kind":"pith_short_16","alias_value":"7ZQDKRBOIDKM4QED","created_at":"2026-07-05T11:25:22.998452+00:00"},{"alias_kind":"pith_short_8","alias_value":"7ZQDKRBO","created_at":"2026-07-05T11:25:22.998452+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/7ZQDKRBOIDKM4QEDZIQ5FHZCTH","json":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH.json","graph_json":"https://pith.science/api/pith-number/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/graph.json","events_json":"https://pith.science/api/pith-number/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/events.json","paper":"https://pith.science/paper/7ZQDKRBO"},"agent_actions":{"view_html":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH","download_json":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH.json","view_paper":"https://pith.science/paper/7ZQDKRBO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.17772&json=true","fetch_graph":"https://pith.science/api/pith-number/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/graph.json","fetch_events":"https://pith.science/api/pith-number/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/action/storage_attestation","attest_author":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/action/author_attestation","sign_citation":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/action/citation_signature","submit_replication":"https://pith.science/pith/7ZQDKRBOIDKM4QEDZIQ5FHZCTH/action/replication_record"}},"created_at":"2026-07-05T11:25:22.998452+00:00","updated_at":"2026-07-05T11:25:22.998452+00:00"}