AI ‘breakthrough’: neural net has human-like ability to generalize language

https://www.nature.com/articles/d41586-023-03272-3

28 points · 9 comments · view on lemmy.world

9 Comments

ASeriesOfPoorChoices@lemmy.world · 16 pts · 2y (5 replies)

therandoe@kbin.social · 4 pts · 2y

That image suddenly became very different. Specifically what you can't see.

abobla@lemm.ee · 3 pts · 2y

damn! He's eating her brain 😭

smort@lemmy.world · 2 pts · 2y (2 replies)

I don’t get it. A little help?

fettuccinecode@programming.dev · 6 pts · 2y (1 reply)

Just compare it to the one from link preview.

smort@lemmy.world · 1 pts · 2y

Oh lol thanks

vcmj@programming.dev · 5 pts · 2y (1 reply)

Anybody have a link to the paper? The article strikes me as a used car salemans trying to sell me a journal. Mostly what I'm getting is new reinforcement learning technique catered to language? But what model architecture? Is it new? I'd like to know

vcmj@programming.dev · 4 pts · 2y

Found it in a cross post: https://www.nature.com/articles/s41586-023-06668-3

Its a transformer, someone fire the journalist. Still, interesting stuff

NerdyPopRocks@programming.dev · 3 pts · 2y

Comparing to a study of 25 humans seems like too small a sample to say it performs as well as humans generally