SWE-Touch：当用户修改代码时编码代理的基准测试
原标题：SWE-Touch: Benchmarking Coding Agents When Users Touch the Code
AI 摘要
SWE-Touch是一个新框架，用于在共享工作空间中测试编码代理，通过注入与任务冲突的用户编辑来模拟真实开发场景。在SWE-bench Verified上，Counter-Edit使平均解决率下降7.7个百分点，并重新洗牌了模型排名。Claude Opus 4.8和GPT 5.5表现出较强的韧性，而开源模型在自主基准上表现良好但在协作场景中显著退化。结果表明，当前模型缺乏检测工作区变化、协调冲突编辑和验证受影响行为的能力。
正文节选
SWE-Touch: Benchmarking Coding Agents When Users Touch the Code Abstract Real-world software development requires coding agents to operate in shared workspaces where users may inspect and modify code during an ongoing task, yet existing repository-level benchmarks typically evaluate agents working alone or restrict user participation to messages. This leads us to ask: how do coding agents understand and respond to code changes in a shared workspace? We introduce SWE-Touch, a framework that stres