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 maxdoop on Monday October 09, 2023 @04:51PM (#37825337)
We know exactly what an LLM does? How does that differ from a brain?
[ Parent | on HN ]
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by RC_ITR on Monday October 09, 2023 @05:32PM (#37825794)
One does deterministic matrix math and the other doesn’t.

It’s a common experience that candy tasted sweeter to you as a child, right?

How does that work in a transformer?

[ Parent | on HN ]