标题： SWE-Perf：语言模型能否在现实世界代码库上优化代码性能？ 显示英文标题

标题： SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?

摘要： 代码性能优化在现实世界的软件工程中至关重要，对生产级系统尤为关键。 虽然大型语言模型（LLMs）在代码生成和错误修复方面表现出色，但它们在仓库级别提升代码性能的能力仍鲜有研究。 为解决这一差距，我们引入了SWE-Perf，这是首个专门设计用于在真实仓库环境中系统评估LLMs在代码性能优化任务上的基准。 SWE-Perf包含140个精心挑选的实例，每个实例均来自流行GitHub仓库中的性能改进拉取请求。 每个基准实例包括相关代码库、目标函数、与性能相关的测试、专家编写的补丁和可执行环境。 通过对涵盖文件级别和仓库级别方法的代表性方法（例如Agentless和OpenHands）进行全面评估，我们揭示了现有LLMs与专家级优化性能之间的显著能力差距，突显了这一新兴领域中的关键研究机会。 显示英文摘要

摘要： Code performance optimization is paramount in real-world software engineering and critical for production-level systems. While Large Language Models (LLMs) have demonstrated impressive capabilities in code generation and bug fixing, their proficiency in enhancing code performance at the repository level remains largely unexplored. To address this gap, we introduce SWE-Perf, the first benchmark specifically designed to systematically evaluate LLMs on code performance optimization tasks within authentic repository contexts. SWE-Perf comprises 140 carefully curated instances, each derived from performance-improving pull requests from popular GitHub repositories. Each benchmark instance includes the relevant codebase, target functions, performance-related tests, expert-authored patches, and executable environments. Through a comprehensive evaluation of representative methods that span file-level and repo-level approaches (e.g., Agentless and OpenHands), we reveal a substantial capability gap between existing LLMs and expert-level optimization performance, highlighting critical research opportunities in this emerging field.