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

arXiv: https://arxiv.org/abs/2604.05963 · HTML: https://mlsanity.com/paper/2604.05963/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 22 references that are in the map.
- First posted on arXiv 7 Apr 2026.

## Builds on (most cited first, 14 of 14)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Agentless: Demystifying LLM-based Software Engineering Agents](https://mlsanity.com/paper/2407.01489/) (2024)
- [Speculative Decoding: Exploiting Speculative Execution for Accelerating Seq2seq Generation](https://mlsanity.com/paper/2203.16487/) (2022)
- [VerilogCoder: Autonomous Verilog Coding Agents with Graph-based Planning and Abstract Syntax Tree (AST)-based Waveform Tracing Tool](https://mlsanity.com/paper/2408.08927/) (2024)
- [LeetCodeDataset: A Temporal Dataset for Robust Evaluation and Efficient Training of Code LLMs](https://mlsanity.com/paper/2504.14655/) (2025)
- [Interactive Debugging and Steering of Multi-Agent AI Systems](https://mlsanity.com/paper/2503.02068/) (2025)
- [QiMeng-CodeV-R1: Reasoning-Enhanced Verilog Generation](https://mlsanity.com/paper/2505.24183/) (2025)
- [Automatic Program Repair with OpenAI's Codex: Evaluating QuixBugs](https://mlsanity.com/paper/2111.03922/) (2021)
- [Aligning the Objective of LLM-based Program Repair](https://mlsanity.com/paper/2404.08877/) (2024)
- [HLSDebugger: Identification and Correction of Logic Bugs in HLS Code with LLM Solutions](https://mlsanity.com/paper/2507.21485/) (2025)
- [Towards LLM-based Root Cause Analysis of Hardware Design Failures](https://mlsanity.com/paper/2507.06512/) (2025)
- [SLMFix: Leveraging Small Language Models for Domain Specific Language Error Fixing with Reinforcement Learning](https://mlsanity.com/paper/2511.19422/) (2025)

## Built on by (most cited first, 2 of 2)

- [When Models Edit Too Much: On the Fidelity of Minimal Code Edits](https://mlsanity.com/paper/2609.04061/) (2026)
- [PreciseCoder: Faithful and Generalizable Debugging Beyond Unit-Test Rewards](https://mlsanity.com/paper/2610.18755/) (2026)

## Abstract (arXiv)

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 fix1@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.