Google DeepMind 推出 Dream-RSI：让 AI 智能体通过“做梦”回放改进搜索策略
原标题：Google Deepmind's Dream-RSI helps AI agents improve by “dreaming” about past attempts
AI 摘要
Google 与 DeepMind 研究人员提出 Dream-RSI 方法，通过回放已完成的搜索记录来测试新的搜索策略，而无需重复昂贵的实时计算，从而让 AI 智能体更高效地解决困难搜索任务。该方法只改变搜索策略、不改动底层模型，在 Gemini 3.1 Pro 和 Gemini 3.7 Flash 上测试八项任务，平均运行时间和尝试次数均下降，并在部分 GPU 任务上以更少运行次数达到相当或更好性能。研究还发现，将搜索历史压缩为明确指令反而可能限制探索，相关代码已在 GitHub 公开。
正文节选
Google Deepmind's Dream-RSI helps AI agents improve by “dreaming” about past attempts Researchers at Google and Deepmind have developed a method that helps AI agents tackle difficult search tasks more efficiently. It uses past search runs to test new strategies without repeating costly computations. Self-improving AI agents are supposed to one day discover new algorithms, solutions to math problems, or faster code on their own. They follow the same basic process of proposing a solution, evaluati