UniDebugger: Hierarchical Multi-Agent Framework for Unified Software Debugging

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lee, Cheryl, Xia, Chunqiu Steven, Yang, Longji, Huang, Jen-tse, Zhu, Zhouruixin, Zhang, Lingming, Lyu, Michael R.
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914162162008064
author Lee, Cheryl
Xia, Chunqiu Steven
Yang, Longji
Huang, Jen-tse
Zhu, Zhouruixin
Zhang, Lingming
Lyu, Michael R.
author_facet Lee, Cheryl
Xia, Chunqiu Steven
Yang, Longji
Huang, Jen-tse
Zhu, Zhouruixin
Zhang, Lingming
Lyu, Michael R.
contents Software debugging is a time-consuming endeavor involving a series of steps, such as fault localization and patch generation, each requiring thorough analysis and a deep understanding of the underlying logic. While large language models (LLMs) demonstrate promising potential in coding tasks, their performance in debugging remains limited. Current LLM-based methods often focus on isolated steps and struggle with complex bugs. In this paper, we propose the first end-to-end framework, FixAgent, for unified debugging through multi-agent synergy. It mimics the entire cognitive processes of developers, with each agent specialized as a particular component of this process rather than mirroring the actions of an independent expert as in previous multi-agent systems. Agents are coordinated through a three-level design, following a cognitive model of debugging, allowing adaptive handling of bugs with varying complexities. Experiments on extensive benchmarks demonstrate that FixAgent significantly outperforms state-of-the-art repair methods, fixing 1.25$\times$ to 2.56$\times$ bugs on the repo-level benchmark, Defects4J. This performance is achieved without requiring ground-truth root-cause code statements, unlike the baselines. Our source code is available on https://github.com/AcceptePapier/UniDebugger.
format Preprint
id arxiv_https___arxiv_org_abs_2404_17153
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle UniDebugger: Hierarchical Multi-Agent Framework for Unified Software Debugging
Lee, Cheryl
Xia, Chunqiu Steven
Yang, Longji
Huang, Jen-tse
Zhu, Zhouruixin
Zhang, Lingming
Lyu, Michael R.
Software Engineering
Software debugging is a time-consuming endeavor involving a series of steps, such as fault localization and patch generation, each requiring thorough analysis and a deep understanding of the underlying logic. While large language models (LLMs) demonstrate promising potential in coding tasks, their performance in debugging remains limited. Current LLM-based methods often focus on isolated steps and struggle with complex bugs. In this paper, we propose the first end-to-end framework, FixAgent, for unified debugging through multi-agent synergy. It mimics the entire cognitive processes of developers, with each agent specialized as a particular component of this process rather than mirroring the actions of an independent expert as in previous multi-agent systems. Agents are coordinated through a three-level design, following a cognitive model of debugging, allowing adaptive handling of bugs with varying complexities. Experiments on extensive benchmarks demonstrate that FixAgent significantly outperforms state-of-the-art repair methods, fixing 1.25$\times$ to 2.56$\times$ bugs on the repo-level benchmark, Defects4J. This performance is achieved without requiring ground-truth root-cause code statements, unlike the baselines. Our source code is available on https://github.com/AcceptePapier/UniDebugger.
title UniDebugger: Hierarchical Multi-Agent Framework for Unified Software Debugging
topic Software Engineering
url https://arxiv.org/abs/2404.17153