重新思考自动程序修复：缺陷复杂度、缺陷定位与大语言模型成本效率的影响
Rethinking Automated Program Repair: The Impact of Bug Complexity, Fault Localization, and LLM Cost-efficiency
- Colorado State University（科罗拉多州立大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究通过实证分析，发现中等复杂度缺陷可超50%由低成本LLM修复，不精确缺陷定位会扩大APR技术差距，高成本LLM与强推理设置并非总能提升成本效率，DeepSeek-V3.2成本效率最优。
AI中文摘要：
背景：软件缺陷仍是开发中的关键挑战，需要有效的自动程序修复（APR）技术。基于大语言模型（LLM）的APR系统虽已展现潜力，但现有研究主要关注整体修复效果，缺陷复杂度、缺陷定位、推理设置及修复成本效率的影响仍未得到充分探索。目标：本研究针对基于LLM的APR开展全面实证分析，重点探究缺陷复杂度、缺陷定位、推理设置及成本如何影响修复性能。方法：我们使用三种LLM（DeepSeek、GPT和Llama）评估两种APR技术（ChatRepair和CodeCorrector），并通过多维度实证框架与统计分析，考察其在不同缺陷复杂度水平及定位策略下的性能。结果：尽管结构复杂的缺陷与不精确的缺陷定位会增加修复难度，但基于LLM的APR技术仍能达到具有竞争力的修复效果；不精确的缺陷定位会显著扩大APR技术间的性能差距；此外，更高成本的LLM与更强的推理设置并不总能带来更优的成本效率，揭示了修复效果与计算成本之间存在不容忽视的权衡关系。结论：超过50%的中等复杂度缺陷可由低成本基于LLM的APR技术修复；随着缺陷定位精度降低，APR技术间的修复效果差距会扩大；GPT-5比DeepSeek-V4-pro和DeepSeek-V3.2分别多修复7个和39个复杂缺陷，而DeepSeek-V3.2的总修复成本展现出最优的成本效率表现。
英文摘要：
Background: Software bugs remain a critical challenge in development, necessitating effective Automated Program Repair (APR) techniques. While Large Language Model (LLM)-based APR systems have shown promise, prior studies primarily focus on overall repair effectiveness. The effects of bug complexity, fault localization, reasoning settings, and repair cost-effectiveness remain insufficiently explored. Aims: This study presents a comprehensive empirical analysis of LLM-based APR, focusing on how repair performance is shaped by bug complexity, fault localization, reasoning settings, and costs. Method: We evaluate two APR techniques (ChatRepair and CodeCorrector) using three LLMs (DeepSeek, GPT, and Llama), and examine their performance across diverse levels of bug complexity and localization strategies through a multi-dimensional empirical framework and statistical analysis. Results: Although structurally complex bugs and imprecise fault localization make repair more challenging, LLM-based APR techniques still achieve competitive repair effectiveness. Imprecise fault localization can substantially enlarge the performance gap between APR techniques. Furthermore, higher-cost LLMs and stronger reasoning settings do not consistently yield better cost-efficiency, revealing a nontrivial trade-off between repair effectiveness and computational cost. Conclusions: Over 50% of moderately complex bugs can be repaired by low-cost LLM-based APR techniques. The repair effectiveness gap between APR techniques becomes larger as fault localization becomes less precise. GPT-5 repairs 7 and 39 more complex bugs than DeepSeek-V4-pro and DeepSeek-V3.2, respectively; whereas the total repair cost of DeepSeek-V3.2 shows the best cost-efficiency performance.
