跳到正文
The Decoder· Jonathan Kemper·· 21 天前AI 评分60

Google 研究者提出 RRSI,防止自我改进智能体记住测试任务

Google researchers find a way to keep self-improving AI agents from memorizing their tests

AI 导读

Google 研究者发表论文提出 RRSI(Regularized Recursive Self-Improvement of Agent Harnesses),通过随时间收缩的编辑预算和严格 critic 审查,阻止智能体在自我优化中记住测试任务。

来源:The Decoder · the-decoder.com