{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VTPOX5ZNBHOMPBPMFNEZTCWWLI","short_pith_number":"pith:VTPOX5ZN","schema_version":"1.0","canonical_sha256":"acdeebf72d09dcc785ec2b49998ad65a2691f2b9a7b9c64c949afb8ac61b3475","source":{"kind":"arxiv","id":"2602.07457","version":2},"attestation_state":"computed","paper":{"title":"Pull Requests as a Training Signal for Repo-Level Code Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.SE","authors_text":"Lei Ji, Lin Gui, Murong Ma, Peng Cheng, Qinglin Zhu, Runcong Zhao, Shuai Lu, Tianyu Chen, Xiangxiang Dai, Yeyun Gong, Yulan He","submitted_at":"2026-02-07T09:22:25Z","abstract_excerpt":"Repository-level code editing requires models to understand complex dependencies and execute precise multi-file modifications across a large codebase. While recent gains on SWE-bench rely heavily on complex agent scaffolding, it remains unclear how much of this capability can be internalised via high-quality training signals. To address this, we propose Clean Pull Request (Clean-PR), a mid-training paradigm that leverages real-world GitHub pull requests as a training signal for repository-level editing. We introduce a scalable pipeline that converts noisy pull request diffs into Search/Replace"},"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":"2602.07457","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-02-07T09:22:25Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"06d9566f9bc3ba6b53b176066e61e436ea47ce6ac2e50c8f66631d88de2f663e","abstract_canon_sha256":"9730115e7db0b34873c3fc817e30d507f9b80d901e48a46e69d2c83f3dc21f64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:32.573523Z","signature_b64":"ibMDK9Zg9bpvCH1ZOGlfSyZdI0TCU0OP8U96rIQ9g3ll1A5fzgGXeXgBmXdty5MHewvrPtqSnJw1LNcxAfekCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"acdeebf72d09dcc785ec2b49998ad65a2691f2b9a7b9c64c949afb8ac61b3475","last_reissued_at":"2026-06-01T01:02:32.572630Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:32.572630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Pull Requests as a Training Signal for Repo-Level Code Editing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.SE","authors_text":"Lei Ji, Lin Gui, Murong Ma, Peng Cheng, Qinglin Zhu, Runcong Zhao, Shuai Lu, Tianyu Chen, Xiangxiang Dai, Yeyun Gong, Yulan He","submitted_at":"2026-02-07T09:22:25Z","abstract_excerpt":"Repository-level code editing requires models to understand complex dependencies and execute precise multi-file modifications across a large codebase. While recent gains on SWE-bench rely heavily on complex agent scaffolding, it remains unclear how much of this capability can be internalised via high-quality training signals. To address this, we propose Clean Pull Request (Clean-PR), a mid-training paradigm that leverages real-world GitHub pull requests as a training signal for repository-level editing. We introduce a scalable pipeline that converts noisy pull request diffs into Search/Replace"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07457","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/2602.07457/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":"2602.07457","created_at":"2026-06-01T01:02:32.572736+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.07457v2","created_at":"2026-06-01T01:02:32.572736+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07457","created_at":"2026-06-01T01:02:32.572736+00:00"},{"alias_kind":"pith_short_12","alias_value":"VTPOX5ZNBHOM","created_at":"2026-06-01T01:02:32.572736+00:00"},{"alias_kind":"pith_short_16","alias_value":"VTPOX5ZNBHOMPBPM","created_at":"2026-06-01T01:02:32.572736+00:00"},{"alias_kind":"pith_short_8","alias_value":"VTPOX5ZN","created_at":"2026-06-01T01:02:32.572736+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/VTPOX5ZNBHOMPBPMFNEZTCWWLI","json":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI.json","graph_json":"https://pith.science/api/pith-number/VTPOX5ZNBHOMPBPMFNEZTCWWLI/graph.json","events_json":"https://pith.science/api/pith-number/VTPOX5ZNBHOMPBPMFNEZTCWWLI/events.json","paper":"https://pith.science/paper/VTPOX5ZN"},"agent_actions":{"view_html":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI","download_json":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI.json","view_paper":"https://pith.science/paper/VTPOX5ZN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.07457&json=true","fetch_graph":"https://pith.science/api/pith-number/VTPOX5ZNBHOMPBPMFNEZTCWWLI/graph.json","fetch_events":"https://pith.science/api/pith-number/VTPOX5ZNBHOMPBPMFNEZTCWWLI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI/action/storage_attestation","attest_author":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI/action/author_attestation","sign_citation":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI/action/citation_signature","submit_replication":"https://pith.science/pith/VTPOX5ZNBHOMPBPMFNEZTCWWLI/action/replication_record"}},"created_at":"2026-06-01T01:02:32.572736+00:00","updated_at":"2026-06-01T01:02:32.572736+00:00"}