{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GTLCHJ4VOWKPHKEYKOOBCWXXUT","short_pith_number":"pith:GTLCHJ4V","schema_version":"1.0","canonical_sha256":"34d623a7957594f3a898539c115af7a4d479cc6d4c892292e3125e4209482280","source":{"kind":"arxiv","id":"2508.01550","version":2},"attestation_state":"computed","paper":{"title":"RepoForge: Training a SOTA Fast-thinking SWE Agent with an End-to-End Data Curation Pipeline Synergizing SFT and RL at Scale","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Aaditya Bhatia, Ahmed E. Hassan, Arthur Leung, Boyuan Chen, Chengzong Zhao, Chong Chun Yong, Dayi Lin, Gopi Krishnan Rajbahadur, Gustavo A. Oliva, Haoxiang Zhang, Yihao Chen, Zhilong Chen","submitted_at":"2025-08-03T02:34:16Z","abstract_excerpt":"Training software engineering (SWE) LLMs is bottlenecked by expensive infrastructure, inefficient evaluation pipelines, scarce training data, and costly quality control. We present RepoForge, an autonomous, end-to-end pipeline that generates, evaluates, and trains SWE agents at scale. Our key contributions include: (1) RepoForge-8B-Agent, achieving 17.4\\% on SWE-Bench-Verified~\\citep{swebench_verified2024}, establishing new state-of-the-art for $\\leq$8B non-thinking LLMs; (2) 7,304 executable environments auto-generated from real GitHub commits with zero manual intervention; (3) 14$\\times$ sto"},"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":"2508.01550","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.SE","submitted_at":"2025-08-03T02:34:16Z","cross_cats_sorted":[],"title_canon_sha256":"9c40abbc351a6f25d12f5b60ebe7f3b796c8611df684aef5c439eecc747b35eb","abstract_canon_sha256":"695d0e8a0d4c7bfd709f18a018ce0a4d6bd00921de8be41758392a0c1a850c89"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:16.619933Z","signature_b64":"ksxgieQq3zRwyOxjAymXZazPo9SEyBXA5uFBNMkgWRfv9wAFwtq3bdGClBtQAU/7Hk1HOsQt64H9FyanMVIbDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34d623a7957594f3a898539c115af7a4d479cc6d4c892292e3125e4209482280","last_reissued_at":"2026-07-05T12:04:16.619372Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:16.619372Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RepoForge: Training a SOTA Fast-thinking SWE Agent with an End-to-End Data Curation Pipeline Synergizing SFT and RL at Scale","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Aaditya Bhatia, Ahmed E. Hassan, Arthur Leung, Boyuan Chen, Chengzong Zhao, Chong Chun Yong, Dayi Lin, Gopi Krishnan Rajbahadur, Gustavo A. Oliva, Haoxiang Zhang, Yihao Chen, Zhilong Chen","submitted_at":"2025-08-03T02:34:16Z","abstract_excerpt":"Training software engineering (SWE) LLMs is bottlenecked by expensive infrastructure, inefficient evaluation pipelines, scarce training data, and costly quality control. We present RepoForge, an autonomous, end-to-end pipeline that generates, evaluates, and trains SWE agents at scale. Our key contributions include: (1) RepoForge-8B-Agent, achieving 17.4\\% on SWE-Bench-Verified~\\citep{swebench_verified2024}, establishing new state-of-the-art for $\\leq$8B non-thinking LLMs; (2) 7,304 executable environments auto-generated from real GitHub commits with zero manual intervention; (3) 14$\\times$ sto"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.01550","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/2508.01550/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":"2508.01550","created_at":"2026-07-05T12:04:16.619436+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.01550v2","created_at":"2026-07-05T12:04:16.619436+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.01550","created_at":"2026-07-05T12:04:16.619436+00:00"},{"alias_kind":"pith_short_12","alias_value":"GTLCHJ4VOWKP","created_at":"2026-07-05T12:04:16.619436+00:00"},{"alias_kind":"pith_short_16","alias_value":"GTLCHJ4VOWKPHKEY","created_at":"2026-07-05T12:04:16.619436+00:00"},{"alias_kind":"pith_short_8","alias_value":"GTLCHJ4V","created_at":"2026-07-05T12:04:16.619436+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.08088","citing_title":"EnvX: Agentize Everything with Agentic AI","ref_index":2025,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT","json":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT.json","graph_json":"https://pith.science/api/pith-number/GTLCHJ4VOWKPHKEYKOOBCWXXUT/graph.json","events_json":"https://pith.science/api/pith-number/GTLCHJ4VOWKPHKEYKOOBCWXXUT/events.json","paper":"https://pith.science/paper/GTLCHJ4V"},"agent_actions":{"view_html":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT","download_json":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT.json","view_paper":"https://pith.science/paper/GTLCHJ4V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.01550&json=true","fetch_graph":"https://pith.science/api/pith-number/GTLCHJ4VOWKPHKEYKOOBCWXXUT/graph.json","fetch_events":"https://pith.science/api/pith-number/GTLCHJ4VOWKPHKEYKOOBCWXXUT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT/action/storage_attestation","attest_author":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT/action/author_attestation","sign_citation":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT/action/citation_signature","submit_replication":"https://pith.science/pith/GTLCHJ4VOWKPHKEYKOOBCWXXUT/action/replication_record"}},"created_at":"2026-07-05T12:04:16.619436+00:00","updated_at":"2026-07-05T12:04:16.619436+00:00"}