CrossSWE: 测试跨仓库联合开发的大语言模型基准
CrossSWE: Benchmarking Cross-Repository Joint Development for Coding Agents
CrossSWE 基准测试当前大语言模型在跨仓库联合软件开发中的跨仓库一致性，包含 171 个实例，涵盖 165 个仓库、20 个技术领域和 13 种编程语言，评估显示最强的 GPT-6.1 Sol 仅达 58.48% Pass@1，跨仓库不一致是主要失败模式。
Published on Nov 1
Authors:
,
,
,
,
,
Abstract
CrossSWE introduces a multi-repository benchmark showing that current LLMs struggle with cross-repository consistency during joint software development.
As coding agents achieve increasingly high success rates on software engineering benchmarks, recent efforts have sought to increase task complexity through longer horizons and more extensive code modifications. However, real-world software development often spans multiple interdependent repositories, introducing additional challenges in dependency reasoning, change propagation, and interface consistency. Rather than further increasing complexity within individual repositories, we introduce CrossSWE, a benchmark for cross-repository joint development comprising 171 real-world instances spanning 165 repositories, 20 technical domains, and 13 programming languages. CrossSWE verifies the necessity of coordinated changes by testing whether partial repository modifications are insufficient to achieve the intended behavior. Our evaluation of five LLMs reveals substantial challenges, with the strongest model, GPT-6.1 Sol, achieving only 58.48% Pass@1. Furthermore, cross-repository inconsistencies emerge as the most prevalent failure mode (44.2%), highlighting the difficulty of maintaining consistency across interdependent repositories.
View arXiv page View PDF Project page Add to collection
Get this paper in your agent:
hf papers read 2611.01484
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2611.01484 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2611.01484 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2611.01484 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
来源: Hugging Face Daily Papers · huggingface.co