News for hackers. Stuff that matters.
Re:LLMs are not greater than the sum of their parts: Stanfor... by stevenhuang on Tuesday May 09, 2023 @04:39AM (#35871628)
How can it not be. Like think about what you're saying here.

Would you rather be able to evaluate a model on it's demonstrated capabilities (multistep reasoning, question and answer, instruction following, theory of mind, etc) or some nebulous metric along an axis that may as well not correspond to practice.

We only care about how good AI is at things that matter to us as humans.

There are no comments on this comment yet.