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 stavros on Monday October 09, 2023 @03:40PM (#37824489)
The difference is that your goldfish doesn't look like it's walking. I'm sure my door doesn't think. I'm not sure the LLM doesn't.
[ Parent | on HN ]
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by antisthenes on Monday October 09, 2023 @04:46PM (#37825284)
> I'm not sure the LLM doesn't.

If you turn off the power to the computer hosting the LLM, you can also be sure it doesn't think.

[ Parent | on HN ]