Can LLMs Demystify Bug Reports?

TL;DR

This paper investigates whether ChatGPT can understand and reproduce software bugs from natural language reports, demonstrating that it can successfully address about half of the bugs in a standard benchmark, indicating promising potential for automated bug handling.

Contribution

The study evaluates ChatGPT's ability to understand and reproduce bug reports, revealing its potential to automate part of the bug fixing process with minimal human intervention.

Findings

ChatGPT successfully reproduces 50% of bugs in Defects4J.

LLMs show promise in automating bug analysis and reproduction.

Potential for reducing manual effort in bug fixing processes.

Abstract

Bugs are notoriously challenging: they slow down software users and result in time-consuming investigations for developers. These challenges are exacerbated when bugs must be reported in natural language by users. Indeed, we lack reliable tools to automatically address reported bugs (i.e., enabling their analysis, reproduction, and bug fixing). With the recent promises created by LLMs such as ChatGPT for various tasks, including in software engineering, we ask ourselves: What if ChatGPT could understand bug reports and reproduce them? This question will be the main focus of this study. To evaluate whether ChatGPT is capable of catching the semantics of bug reports, we used the popular Defects4J benchmark with its bug reports. Our study has shown that ChatGPT was able to demystify and reproduce 50% of the reported bugs. ChatGPT being able to automatically address half of the reported bugs shows promising potential in the direction of applying machine learning to address bugs with only a human-in-the-loop to report the bug.