Claude Opus 5.5 操控机械臂难判任务终止
Knowing when to stop is hard for language models controlling a robot arm
AI 导读
Artificial Analysis 指出，语言模型操控机械臂时难以判断何时停止。Claude Opus 5.5 用 23 分钟完成任务并称“鳄鱼看起来已经完全进去了”，随后又花 22 分钟来回推鳄鱼，最终几乎回到原位。
正文
Knowing when to stop is hard for language models controlling a robot arm
➤ Claude Opus 5.5 completed the task in 23 minutes, and said so: "The crocodile looks fully inside now"
➤ It then spent 22 minutes pushing the crocodile back and forth, only to finish almost exactly where it was before