SWE-Bench ProMax：多语言大规模代码重构基准
原标题：SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring
AI 摘要
Hugging Face 发布 SWE-Bench ProMax，一个由专家策划的多语言代码重构基准，包含 170 个实例，覆盖七种编程语言。该基准通过重写问题描述和人工审查测试套件，解决了现有基准中测试缺陷和训练数据泄漏的问题。实验显示，前沿模型在两种代理框架下的最佳解决率仅为 41.2%，表明该基准对当前 AI 编码代理构成有意义的挑战。基准已在 Hugging Face 上公开。
正文节选
SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code Refactoring Abstract As AI coding agents take on increasingly complex, long-horizon software engineering tasks, existing benchmarks are rapidly saturating and their evaluation quality has come under serious scrutiny: a recent audit found that nearly 60% of unsolved SWE-bench Verified instances contain flawed tests -- either overly narrow tests that reject correct solutions or overly broad tests that check unstated requirement