SWE-Factory: Your Automated Factory for Issue Resolution Training Data and Evaluation Benchmarks

TL;DR

SWE-Factory is an automated pipeline that constructs large-scale GitHub issue resolution datasets, addressing previous manual limitations and enabling efficient evaluation of LLMs in software engineering tasks.

Contribution

It introduces SWE-Factory, a fully automated data construction pipeline with novel components for binary file recovery, LLM-based environment setup, and log parsing for validation.

Findings

Successfully constructed evaluation datasets for 671 issues across four languages.

Achieved high validation accuracy with an F1 score of 0.99.

Demonstrated cost-effective environment construction at $0.047 per instance.

Abstract

Constructing large-scale datasets for the GitHub issue resolution task is crucial for both training and evaluating the software engineering capabilities of Large Language Models (LLMs). However, the existing GitHub issue resolution data construction pipeline is challenging and labor-intensive. We identify three key limitations in existing pipelines: (1) test patches collected often omit binary file changes; (2) the manual construction of evaluation environments is labor-intensive; and (3) the fail2pass validation phase requires manual inspection of test logs and writing custom parsing code to extract test status from logs. In this paper, we propose SWE-Factory, a fully automated issue resolution data construction pipeline, to resolve these limitations. First, our pipeline automatically recovers missing binary test files and ensures the correctness of test patches. Second, we introduce SWE-Builder, a LLM-based multi-agent system that automates evaluation environment construction. Third, we introduce a standardized, exit-code-based log parsing method to automatically extract test status, enabling a fully automated fail2pass validation. Experiments on 671 real-world GitHub issues across four programming languages show that our method can effectively construct valid evaluation environments for GitHub issues at a reasonable cost. For example, with GPT-4.1 mini, our SWE-Builder constructs 337 valid task instances out of 671 issues, at $0.047 per instance. Our ablation study further shows the effectiveness of different components of SWE-Builder. We also demonstrate through manual inspection that our exit-code-based fail2pass validation method is highly accurate, achieving an F1 score of 0.99. Additionally, we conduct an exploratory experiment to investigate whether we can use SWE-Factory to enhance models' software engineering ability.