一句话：提出智能体框架 AEGIS，缓解检索信息冗长与复现脚本修改缺乏引导的难题，提升从 issue 描述自动复现缺陷的效果。
从 issue 描述中自动复现缺陷能帮助开发者及时定位并修复问题，对软件开发与维护意义重大。借助大语言模型（LLM）强大的理解能力，基于智能体的方法在该任务上取得了当前最优表现：通常以 LLM 作为中央控制器，先检索与缺陷相关的信息作为上下文，再生成缺陷复现脚本，并在脚本修改过程中反复迭代，直到执行信息能准确反映缺陷或迭代轮次耗尽。然而这类方法仍面临两个挑战：（1）检索到的缺陷相关信息篇幅冗长且包含无关片段，LLM 难以充分理解；（2）复现脚本的生成缺乏引导，往往是随机修改，容易产生重复或虚假的改动，最终导致缺陷复现失败。针对这两点，论文提出基于智能体的框架 AEGIS。受所给摘要篇幅所限，其具体机制与实验评估细节未在文中给出。
Automatically reproducing bugs in issue descriptions helps developers pinpoint and fix bugs timely, greatly facilitating the software development and maintenance. Built upon the powerful understanding capabilities of large language models (LLMs), agent-based approaches have achieved the state-of-the-art performance in the task. They generally leverage LLMs as the central controller to first retrieve bug-related information as context and then generate bug reproduction scripts. During the script modification process, agent-based approaches modify the script iteratively until the execution information reflects the bug accurately or the iterative turns are exhausted. Nevertheless, the agent-based approaches still face the following challenges: (1) Lengthy retrieved bug-related information: The retrieved bug-related information is usually long in length and contains irrelevant snippets, which is hard to be well comprehended by LLMs. (2) Lack of guidance in bug reproduction script generation: They generally modify bug reproduction scripts randomly and tend to generate repeated or spurious modifications, leading to bug reproduction failure.
@inproceedings{paperbot9059,
title = {AEGIS: An Agent-based Framework for Bug Reproduction from Issue Descriptions},
author = {Xinchen Wang and Pengfei Gao and Xiangxin Meng and Chao Peng and Ruida Hu and Yun Lin and Cuiyun Gao},
booktitle = {FSE},
year = {2025},
doi = {10.1145/3696630.3728557}
}