The Tower of Hanoi puzzle is exactly the type of artificial, laboratory-based task that my Naturalistic Decision Making (NDM) community has avoided. There is no expertise. No context, no uncertainty.
Software engineer Sean Goedecke offered a similar critique of the Apple paper on his blog, noting that when faced with Tower of Hanoi requiring over 1,000 moves, DeepSeek-R1 "immediately decides ...
一些您可能无法访问的结果已被隐去。
显示无法访问的结果