The Decoder· Jonathan Kemper·· 4 小时前AI 评分62
Google 研究者提出 RRSI,防止自我改进智能体记忆测试任务
Google researchers find a way to keep self-improving AI agents from memorizing their tests
AI 导读
Google 研究者提出 RRSI(Regularized Recursive Self-Improvement of Agent Harnesses),通过随时间收缩的编辑预算和严格评审机制,阻止自我改进的智能体记忆测试任务。
来源:The Decoder · the-decoder.com