Doesn't that presume that one model dominates the other? | Hacker News

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

		esafak 63 days ago \| parent \| context \| favorite \| on: Task-free intelligence testing of LLMs Doesn't that presume that one model dominates the other?

nestorD 62 days ago [–]

It presumes some models are better than others (and we do find that providing data with a wide mix of model strengths improves convergence) but it does not need to be one model, and it does not even need to be transitive.

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