From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Chenglin, Xu, Yisen, Wang, Zehao, Tan, Shin Hwei, Tse-Hsun, Chen
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910268235186176
author Li, Chenglin
Xu, Yisen
Wang, Zehao
Tan, Shin Hwei
Tse-Hsun
Chen
author_facet Li, Chenglin
Xu, Yisen
Wang, Zehao
Tan, Shin Hwei
Tse-Hsun
Chen
contents Repository-level automated program repair (APR) requires long-horizon reasoning over interdependent decisions. However, most LLM-based approaches reconstruct repair reasoning independently for each issue, failing to reuse successful patterns from prior repairs, even though real-world repositories contain many related issues with shared structure or constraints. Existing methods typically rely on forward exploration, which operates under outcome uncertainty, incurs substantial inference-time overhead, and can drift from the final correct patch. We propose Conditional Reasoning Distillation (ConRAD), which leverages in-repository resolved issues by reconstructing repair reasoning backward from verified patches and distilling outcome-consistent, stage-wise repair reasoning plans. Injected at inference time, these plans guide fault localization and patch generation, replacing open-ended exploration with constrained inference without fine-tuning or search. On SWE-Bench Lite, ConRAD improves Pass@1 by 10.4\% (GPT-4o), 8.6\% (DeepSeek-V3), and 10.3\% (GPT-5), demonstrating a scalable inference-time alternative to forward exploration for long-horizon APR.
format Preprint
id arxiv_https___arxiv_org_abs_2601_23257
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair
Li, Chenglin
Xu, Yisen
Wang, Zehao
Tan, Shin Hwei
Tse-Hsun
Chen
Software Engineering
Repository-level automated program repair (APR) requires long-horizon reasoning over interdependent decisions. However, most LLM-based approaches reconstruct repair reasoning independently for each issue, failing to reuse successful patterns from prior repairs, even though real-world repositories contain many related issues with shared structure or constraints. Existing methods typically rely on forward exploration, which operates under outcome uncertainty, incurs substantial inference-time overhead, and can drift from the final correct patch. We propose Conditional Reasoning Distillation (ConRAD), which leverages in-repository resolved issues by reconstructing repair reasoning backward from verified patches and distilling outcome-consistent, stage-wise repair reasoning plans. Injected at inference time, these plans guide fault localization and patch generation, replacing open-ended exploration with constrained inference without fine-tuning or search. On SWE-Bench Lite, ConRAD improves Pass@1 by 10.4\% (GPT-4o), 8.6\% (DeepSeek-V3), and 10.3\% (GPT-5), demonstrating a scalable inference-time alternative to forward exploration for long-horizon APR.
title From Historical Patches to Repair Plans: Outcome-Conditioned Reasoning for Repository-Level Program Repair
topic Software Engineering
url https://arxiv.org/abs/2601.23257