News for hackers. Stuff that matters.
Re:LLMs can't self-correct in reasoning tasks, DeepMind stud... by jddj on Monday October 09, 2023 @03:48PM (#37824618)
Just to throw the other side's perspective of the argument into the ring, the ones reducing the observed phenomena to obviously anything can seem emotional and defensive of some metaphysical person-ness.

There are no comments on this comment yet.