SWE-bench-java: A GitHub Issue Resolving Benchmark for Java

TL;DR

This paper introduces SWE-bench-java, a new benchmark dataset for evaluating large language models' ability to resolve GitHub issues in Java, expanding the existing Python-focused benchmark to support multilingual programming tasks.

Contribution

The paper presents the first Java version of SWE-bench, including dataset, evaluation environment, and leaderboard, enabling multilingual assessment of issue resolving capabilities.

Findings

Implemented SWE-agent as a baseline method

Tested several powerful LLMs on SWE-bench-java

Dataset and tools are publicly available for community use

Abstract

GitHub issue resolving is a critical task in software engineering, recently gaining significant attention in both industry and academia. Within this task, SWE-bench has been released to evaluate issue resolving capabilities of large language models (LLMs), but has so far only focused on Python version. However, supporting more programming languages is also important, as there is a strong demand in industry. As a first step toward multilingual support, we have developed a Java version of SWE-bench, called SWE-bench-java. We have publicly released the dataset, along with the corresponding Docker-based evaluation environment and leaderboard, which will be continuously maintained and updated in the coming months. To verify the reliability of SWE-bench-java, we implement a classic method SWE-agent and test several powerful LLMs on it. As is well known, developing a high-quality multi-lingual benchmark is time-consuming and labor-intensive, so we welcome contributions through pull requests or collaboration to accelerate its iteration and refinement, paving the way for fully automated programming.