News for hackers. Stuff that matters.
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by wouldbecouldbe on Monday October 09, 2023 @03:23PM (#37824245)
Machines of meaning probably, text generation is more a side effect or maybe a feature, but not the core of our reasoning let alone being.

There are no comments on this comment yet.