News for hackers. Stuff that matters.

A comment on LLMs can't self-correct in reasoning tasks, DeepMind study finds. Parent comment

Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by hackinthebochs on Monday October 09, 2023 @07:01PM (#37826535)
In this[1] piece I argue that LLMs do, in some cases, demonstrate understanding.

[1] https://www.reddit.com/r/naturalism/comments/1236vzf

[ Parent | on HN ]

There are no comments on this comment yet.