And then still people are saying: these are just heuristic next token predictors... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		nuancebydefault on Nov 28, 2024 \| parent \| context \| favorite \| on: QwQ: Alibaba's O1-like reasoning LLM And then still people are saying: these are just heuristic next token predictors incapable of reasoning.

jjeaff on Nov 29, 2024 [–]

What's weird, is that they are heuristic next token predictors. But that prediction, mixed with a little randomness is producing very seemingly "reasoned" responses.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact