Hi, contributor to Entropix here. This is just my opinion, but I don't think it goes counter to the Bitter Lesson at all, because it's meant to leverage model computation capabilities. Several papers have suggested that models internally compute certainty (https://arxiv.org/abs/2406.16254), and in my view our method simply leverages this computation and factors it explicitly into decoding.
This is as opposed to pure sampling + next token prediction which basically randomly chooses a token. So if a model does 1274 x 8275 and it's not very sure of the answer, it still confidently gives an answer even though it's uncertain and needs to do more working.
Re:Detecting when LLMs are uncertain
by danielmarkbruce on Friday October 25, 2024 @05:05PM
(#41949767)
100%. It's in line with bitter lesson learnings. Good going.
Re:Detecting when LLMs are uncertain
by danielmarkbruce on Friday October 25, 2024 @05:04PM
(#41949756)
Yeah i don't think it's counter at all. The bitter lesson calls out the fact that more computation/search wins.
Honestly it goes counter to the Bitter Lesson (http://www.incompleteideas.net/IncIdeas/BitterLesson.html, which stems from getting too fancy about maze traversal in Chess. But at the scale LLMs are at right now, the improvements might be worth it.