Beyond Localization: Recoverable Headroom and Residual Frontier in Repository-Level RAG-APR
Pengtao Zhao, Boyang Yang, Bach Le, Feng Liu, Haoye Tian

TL;DR
This paper investigates the limits of repository-level automated program repair (APR) after strengthening localization, identifying residual gains and the factors influencing repair success across different paradigms.
Contribution
It introduces a protocol to analyze recoverable gains in repository-level RAG-APR and evaluates the impact of localization, context, and interface design on repair outcomes.
Findings
Oracle Localization improves repair success but remains below 50%.
Additional context probes yield diminishing returns after initial gains.
Prompt-level fusion offers limited improvements beyond existing methods.
Abstract
Repository-level automated program repair (APR) increasingly treats stronger localization as the main path to better repair. We ask a more targeted question: once localization is strengthened, which post-localization levers still provide recoverable gains, which are bounded within our protocol, and what residual frontier remains? We study this question on SWE-bench Lite with three representative repository-level RAG-APR paradigms, Agentless, KGCompass, and ExpeRepair. Our protocol combines Oracle Localization, within-pool Best-of-K, fixed-interface added context probes with per-condition same-token filler controls and same-repository hard negatives, and a common-wrapper oracle check. Oracle Localization improves all three systems, but Oracle success still stays below 50%. Extra candidate diversity still helps inside the sampled 10-patch pools, but that headroom saturates quickly. Under…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
