代码大模型 / AI 编程

Enhancing Automated Program Repair through Fine-tuning and Prompt Engineering

Rishov Paul, Md. Mohib Hossain, Mohammed Latif Siddiq, Masum Hasan, Anindya Iqbal, Joanna C. S. Santos

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.LG

Comments 12 pages, 2 figures, 4 tables

KNOD: Domain Knowledge Distilled Tree Decoder for Automated Program Repair

Nan Jiang, Thibaud Lutellier, Yiling Lou, Lin Tan, Dan Goldwasser, Xiangyu Zhang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.AI

Comments This paper is accepted by 2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE)

RunBugRun -- An Executable Dataset for Automated Program Repair

Julian Aron Prenner, Romain Robbes

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.LG

CCRep: Learning Code Change Representations via Pre-Trained Code Model and Query Back

Zhongxin Liu, Zhijie Tang, Xin Xia, Xiaohu Yang

专题命中 程序修复 ：code model(title,abstract)；分类 cs.SE、cs.AI

Comments Accepted by ICSE 2023

Conversational Automated Program Repair

Chunqiu Steven Xia, Lingming Zhang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.LG

A Syntax-Guided Edit Decoder for Neural Program Repair

Qihao Zhu, Zeyu Sun, Yuan-an Xiao, Wenjie Zhang, Kang Yuan, Yingfei Xiong, Lu Zhang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.AI

Comments 10 pages

Journal ref The ACM Joint European Software Engineering Conference and Symposium on the Foundations of Software Engineering (ESEC/FSE 2021). This is the newest version and corrects some errors

Tea: Program Repair Using Neural Network Based on Program Information Attention Matrix

Wenshuo Wang, Chen Wu, Liang Cheng, Yang Zhang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.AI

Towards a Benchmark Set for Program Repair Based on Partial Fixes

Dirk Beyer, Lars Grunske, Thomas Lemberger, Minxing Tang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.PL

Comments 9 pages, 5 figures

Obstacles in Fully Automatic Program Repair: A survey

S. Amirhossein Mousavi, Donya Azizi Babani, Francesco Flammini

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.AI

SequenceR: Sequence-to-Sequence Learning for End-to-End Program Repair

Zimin Chen, Steve Kommrusch, Michele Tufano, Louis-Noël Pouchet, Denys Poshyvanyk, Martin Monperrus

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.LG

Comments 21 pages, 15 figures

Journal ref IEEE Transactions on Software Engineering, 2019

Path-Based Program Repair

Heinz Riener, Rüdiger Ehlers, Görschwin Fey

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE、cs.PL

Comments In Proceedings FESCA 2015, arXiv:1503.04378

Journal ref EPTCS 178, 2015, pp. 22-32

ASAP-Repair: API-Specific Automated Program Repair Based on API Usage Graphs

Sebastian Nielebock, Paul Blockhaus, Jacob Krüger, Frank Ortmeier

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

Comments Accepted for presentation at the 5th ACM/IEEE International Workshop on Automated Program Repair (APR24)

Applying CodeBERT for Automated Program Repair of Java Simple Bugs

Ehsan Mashhadi, Hadi Hemmati

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE；repository(comments)

Comments Added the repository of source code

Hype Meets Reality: Large Language Models as Mutators in Search-based Automated Program Repair of Simulink-Stateflow Models

炒作与现实的碰撞：大型语言模型作为基于搜索的Simulink-Stateflow模型自动化程序修复中的变异器

Ayesha Irshad, Pablo Valle, Jon Ayerdi, Aitor Arrieta

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 本文以Simulink/Stateflow建模的CPS为研究对象，扩展FlowRepair方法用LLM生成的变异替换部分算子，受控实验发现该方法大幅降低修复性能，分析指出盲目集成LLM到基于搜索的APR存在根本局限，需发展混合方法。

OdinEval: A Reproducible Benchmark for LLM-Based Program Repair in the Odin Programming Language

OdinEval：用于基于大语言模型的Odin编程语言程序修复的可复现基准测试

Bang Xie, Hao Liu, Zhiyuan Peng, Xin Yin, Senjian Zhang, Yuan Luo, Chenhao Ying, Haiming Jin, Wei Chen, Shaocong Long, Zhenyu Shi

专题命中 程序修复 ：program repair(title)；repository(abstract)；分类 cs.SE

AI总结 针对Odin语言程序修复的空白，本文构建可复现基准OdinEval，评估6个语言模型，Kimi-K3解决率66.7%、Qwen3.8-Max复现率96.4%，发布配套多类资源。

Comments 8 pages, 3 figures, 2 tables

ORCA: Observability-Grounded Program Repair for Microservice Incidents

ORCA：基于可观测性的微服务故障程序修复

Yuanchen Gao, Yifang Tian, Yiran Li, Charles Zhang, Hans-Arno Jacobsen

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 ORCA是基于可观测性的微服务故障修复流水线，通过遥测数据提炼故障特征定位候选位置，经多智能体生成补丁并通过遥测验证器评估，在575案例基准中成本效益优于所有基线。

How Small is Enough? Empirical Evidence of Quantized Small Language Models for Automated Program Repair

Kazuki Kusama, Honglin Shu, Masanari Kondo, Yasutaka Kamei

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

Journal ref 2025 ACM/IEEE International Symposium on Empirical Software Engineering and Measurement (ESEM), pp. 393-399, 2025

CausalRepair: Bridging the Causality Gap in Large Language Model-Based Automated Program Repair via Dual-Slicing

CausalRepair：通过双重切片弥合基于大语言模型的自动化程序修复中的因果差距

Linhao Wu, Yizhou Chen, Zhen Yang, Pengyu Xue, Dan Hao

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 CausalRepair是基于最小因果上下文的对话驱动型自动化程序修复框架，通过双重切片策略构建因果相关上下文，在Defects4J数据集上修复313个漏洞，性能优于现有方法且修复成本更低。

From Guessing to Seeing: Enhancing LLM-Based Program Repair via Trace-Guided Multi-strategy Debate

通过多智能体辩论引导的运行时执行轨迹指导的自动程序修复

Jiaqing Wu, Tong Wu, Manqing Zhang, Yunwei Dong, Bo Shen

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 本文提出TraceRepair框架，利用运行时事实作为共享约束进行补丁验证，通过多智能体协作提高修复效率和泛化能力，实验显示其在修复缺陷和处理新 bug 数据集上表现优异。

Comments 13 pages, 4 figures, 10 tables. Accepted at the 41st IEEE/ACM International Conference on Automated Software Engineering (ASE 2026)

Adversarial Bug Reports as a Security Risk in Language Model-Based Automated Program Repair

对抗性错误报告：基于语言模型的自动程序修复中的安全风险

Piotr Przymus, Andreas Happe, Jürgen Cito

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 本文研究对抗性错误报告对基于大语言模型的自动程序修复系统的安全威胁，通过实证分析发现现有防御措施不足，并提出自动化生成对抗性错误报告的原型框架。

Enhancing Program Repair with Specification Guidance and Intermediate Behavioral Signals

通过规范引导和中间行为信号增强程序修复

Minh Le-Anh, Cuong Chi Le, Tien N. Nguyen

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 SpecTune通过整合中间行为推理提升程序修复，利用执行检查点分解修复任务，生成局部后置条件以识别程序行为偏差，结合规范验证信号和判别信号提高修复精度和有效性。

PSearch: Search-based Patch Generation in the Era of LLM-based Automated Program Repair

PSearch：基于大语言模型的自动程序修复时代中基于搜索的补丁生成

Haichuan Hu, Ye Shang, Weifeng Sun, Quanjun Zhang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 针对现有基于大语言模型的自动程序修复方法靠试错生成补丁、难以有效评估搜索方向潜力和分配预算的问题，提出Psearch框架，以迭代补丁评估和优化为核心，能集成不同搜索算法，实验证明其有效性。

Comments accepted to ASE 2026

Smaller Models, Unexpected Costs: Trade-offs in LLM Quantization for Automated Program Repair

更小的模型，意外的代价：LLM量化在自动程序修复中的权衡

Fernando Vallecillos-Ruiz, Giordano d'Aloisio, Max Hort, Luca Traini, Antinisca Di Marco, Leon Moonen

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 研究LLM量化在自动程序修复中的效果，发现量化虽减少内存但增加推理时间和能耗，且修复问题集差异大，权衡受模型架构和任务复杂度影响。

Comments Accepted for publication in the Research Papers Track of the 42nd IEEE International Conference on Software Maintenance and Evolution (ICSME 2026), 14-18 September 2026, Benevento, Italy

To Run or Not to Run: Analyzing the Cost-Effectiveness of Code Execution in LLM-Based Program Repair

运行还是不运行：分析基于LLM的程序修复中代码执行的成本效益

Zhihao Lin, Junhua Zhu, Mingyi Zhou, Xin Wang, Zhensu Sun, Renyu Yang, David Lo, Li Li

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 通过分析SWE-bench上的代理轨迹和端到端修复尝试，发现代码执行在LLM程序修复中普遍使用但效益不均，限制执行可节省成本且不影响修复成功率。

Comments Accepted to the ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA 2026). 23 pages

PracRepair: LLM-Empowered Automated Program Repair Inspired by Human-Like Debugging Practices

PracRepair: 受人类调试实践启发的大语言模型赋能自动化程序修复

Yu Cheng, Zhongxin Liu, Zhenchang Xing, Chao Ni, Qing Huang, Xiaoxue Ren

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 提出PracRepair框架，通过构建按需静态-动态上下文、进行问题驱动的故障诊断并迭代细化补丁，利用动态信息提升LLM在程序修复中的效果，在Defects4J和真实世界漏洞上取得最优性能。

HELO-APR: Enhancing Low-Resource Program Repair through Cross-Lingual Knowledge Transfer

HELO-APR：通过跨语言知识迁移增强低资源程序修复

Zhipeng Wang, Boyang Yang, Yidong Wan, Liuye Guo, You Lv, Tao Zheng, Zhuowei Wang, Tieke He

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 提出HELO-APR框架，通过从高资源语言合成低资源语言修复数据并采用课程学习策略，显著提升低资源语言的自动程序修复效果。

HEJ-Robust: A Robustness Benchmark for LLM-Based Automated Program Repair

HEJ-Robust：基于LLM的自动程序修复的鲁棒性基准

Fazle Rabbi, Jinqiu Yang

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 本文提出HEJ-Robust基准，通过八种语义保持的代码转换生成1450个实例，评估五种微调LLM，在多种转换下模型性能下降超50%，揭示当前LLM修复模型缺乏对细微语法变化的鲁棒性。

Reproducible Automated Program Repair Is Hard -- Experiences With the Defects4J Dataset

可重复的自动程序修复是困难的——Defects4J数据集的经验

Adam Krafczyk, Klaus Schmid

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 本文探讨了Defects4J数据集在自动程序修复中的挑战，发现21.6%的缺陷无法重复，7.1%的测试套件不足，提出了更严格的评估框架以解决测试套件问题。

Comments To be published in Proceedings of the 30th International Conference on Evaluation and Assessment in Software Engineering (EASE'26)

DebugRepair: Enhancing LLM-Based Automated Program Repair via Self-Directed Debugging

DebugRepair: 通过自导向调试增强基于大语言模型的自动程序修复

Linhao Wu, Yifei Pei, Zhen Yang, Kainan Li, Zhonghang Lu, Hao Tan, Xiran Lyu, Jia Li, Yizhou Chen, Pengyu Xue, Kunwu Zheng, Dan Hao

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 DebugRepair通过模拟调试收集中间运行时证据，提升基于大语言模型的自动程序修复性能，实验表明其在修复缺陷方面优于现有方法。

MultiMend: Multilingual Program Repair with Context Augmentation and Multi-Hunk Patch Generation

MultiMend: 多语言程序修复与上下文增强及多补丁补丁生成

Reza Gharibi, Mohammad Hadi Sadreddini, Seyed Mostafa Fakhrahmad

专题命中 程序修复 ：program repair(title,abstract)；分类 cs.SE

AI总结 MultiMend通过上下文增强和多补丁生成技术，提升多语言程序修复效率，修复了2227个bug，其中1545个与开发者补丁一致，121个为多补丁bug。

Journal ref Autom Softw Eng 33, 69 (2026)