{"paper":{"title":"SWE-Edit: Rethinking Code Editing for Efficient SWE-Agent","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SWE-Edit splits code editing into a Viewer subagent for on-demand inspection and an Editor subagent for applying changes from plans, freeing the main agent to reason in cleaner context windows.","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Elsie Nallipogu, Jiaxin Pei, Jin Pan, Junjie Hu, Kenan Li, Maoquan Wang, Qirui Jin, Shengyu Fu, Yikai Zhang, Yufan Huang, Yu Kang, Zijian Jin","submitted_at":"2026-04-28T20:35:09Z","abstract_excerpt":"Large language model agents have made strong progress on software engineering, yet current systems suffer from a context coupling problem: the standard code editing interface conflates code inspection, modification planning, and edit execution within a single context window, forcing agents to interleave exploratory viewing with strictly formatted edit generation. Irrelevant context accumulates and edit reliability degrades. We propose SWE-Edit, which decomposes the editing interface into two specialized subagents: a Viewer that extracts task-relevant code on demand, and an Editor that executes"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On SWE-bench Verified, SWE-Edit improves resolved rate by 2.1% while reducing inference cost by 17.9%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That separating Viewer and Editor subagents plus adaptive mode selection will not introduce coordination overhead or new failure modes that cancel the reported gains; the abstract provides no ablation on subagent communication cost or error propagation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SWE-Edit decomposes agent code editing into specialized subagents and adaptive editing modes, raising resolved rate 2.1% and cutting inference cost 17.9% on SWE-bench Verified while releasing a predictive editing benchmark.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SWE-Edit splits code editing into a Viewer subagent for on-demand inspection and an Editor subagent for applying changes from plans, freeing the main agent to reason in cleaner context windows.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f31315da0c42e0e23446fac6deabe0142379b5e2749cdf18284c1e709825a139"},"source":{"id":"2604.26102","kind":"arxiv","version":2},"verdict":{"id":"9c5476bb-556f-4a56-9dda-f9617bdb2538","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T15:42:45.910204Z","strongest_claim":"On SWE-bench Verified, SWE-Edit improves resolved rate by 2.1% while reducing inference cost by 17.9%.","one_line_summary":"SWE-Edit decomposes agent code editing into specialized subagents and adaptive editing modes, raising resolved rate 2.1% and cutting inference cost 17.9% on SWE-bench Verified while releasing a predictive editing benchmark.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That separating Viewer and Editor subagents plus adaptive mode selection will not introduce coordination overhead or new failure modes that cancel the reported gains; the abstract provides no ablation on subagent communication cost or error propagation.","pith_extraction_headline":"SWE-Edit splits code editing into a Viewer subagent for on-demand inspection and an Editor subagent for applying changes from plans, freeing the main agent to reason in cleaner context windows."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.26102/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T03:35:21.599702Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:33:14.295340Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"648689b798548e3a3317f8a06709f9a579cae46a98ff83290f1af9d3caee6bea"},"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"}