Re:Detecting when LLMs are uncertain
by menhguin on Friday October 25, 2024 @04:01PM
(#41949084)
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.
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.