3 ms·
> PS: Not sure if we're talking about the same argument anymore, I'm following the line of "LLMs work/reason the same as humans" not "AI can't perform as good a
by ben_w 2y ago
> PS: Not sure if we're talking about the same argument anymore, I'm following the line of "LLMs work/reason the same as humans" not "AI can't perform as good as humans (now and in the future)"
Yeah, it's hard to keep track :)
I think we're broadly on the same page with all of this, though. (Even if I might be a little more optimistic on tokenised processing of images).