QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization

TL;DR

This paper introduces PRepair, a novel framework for precise code repair that reduces over-editing by optimizing minimal yet correct modifications using edit-aware rewards.

Contribution

It proposes a new approach combining controlled bug injection and edit-aware policy optimization to enhance repair accuracy and efficiency in program repair tasks.

Findings

PRepair improves repair precision by up to 31.4% under fix_1@1.

It significantly increases decoding throughput with speculative editing.

The framework effectively mitigates over-editing in large language model-based code repair.

Abstract

Large Language Models (LLMs) achieve strong program repair performance but often suffer from over-editing, where excessive modifications overwrite correct code and hinder bug localization. We systematically quantify its impact and introduce precise repair task, which maximizes reuse of correct code while fixing only buggy parts. Building on this insight, we propose PRepair, a framework that mitigates over-editing and improves repair accuracy. PRepair has two components: Self-Breaking, which generates diverse buggy programs via controlled bug injection and min-max sampling, and Self-Repairing, which trains models with Edit-Aware Group Relative Policy Optimization (EA-GRPO) using an edit-aware reward to encourage minimal yet correct edits. Experiments show that PRepair improves repair precision by up to 31.4% under $\mathrm{fix}_1@1$, a metric that jointly considers repair correctness and extent, and significantly increases decoding throughput when combined with speculative editing, demonstrating its potential for precise and practical code repair.