清空聊天历史可帮助 AI 智能体找到更优解
原标题：Clearing chat history helps AI agents find better solutions
AI 摘要
微软研究院一项研究发现，长时间运行的 AI 研究智能体即便增加算力也会陷入性能瓶颈，因为累积的对话上下文会把搜索限制在相似的「想法盆地」中。研究者提出 Fork-and-Flush 方法：将智能体分叉为多个并行分支，每个分支继承工作区文件但清空聊天历史。在 13 个长时任务上，该方法在同等算力下将归一化平均分从单次运行的 0.47 提升至 0.78，相对提升 66%，且结果方差更小。
正文节选
Clearing chat history helps AI agents find better solutions Long-running AI research agents can plateau despite additional compute, a Microsoft Research study finds. Accumulated conversation context can constrain their search, keeping them focused on similar solutions instead of exploring better approaches. Microsoft researchers studied what they call "Autoresearch agents," AI systems that repeatedly propose solutions, evaluate them, and refine them based on feedback. Eight independent runs of t