Francois Chollet On LLMs w/ Active Inference

ARC Prize is a $1,000,000+ public competition to beat and open source a solution to the ARC-AGI benchmark.
Hosted by Mike Knoop (Co-founder, Zapier) and François Chollet (Creator of ARC-AGI, Keras).
--
Website: arcprize.org/
Twitter/X: / arcprize
Newsletter: Signup @ arcprize.org/
Discord: / discord
Try your first ARC-AGI tasks: arcprize.org/play

[@MsFerroCat]

im glad somebody came there to share the misconception that most Models that exist nowadays are NOT ai but just an overly complex imitating machine
even tho it can imitate a lot, learning skills is more of a natural thing, yet not described with code, witch is the goal here, witch is so fun to realize that its more like the process of creating artificial life more than anything

[@ARCprize]

Totally! We agree

[@pascalgula]

genius if nobody did that and performance to be validated on my use-case. when will be released the first Open Source component derived from this initiative ? (sorry if I it is already available but didn't find it yet...)

[@ignaciosavi7739]

I advise you to not get that cooky. Otherwise ill solve everything and demand my prize 😂

[@ARCprize]

We're excited to see you on the leaderboard!

[@ignaciosavi7739]

@@ARCprize sheesh ok I'll be there;-)

[@tankieslayer6927]

I am not entirely convinced about active inference. Training on unlabelled test data is an old trick on Kaggle to improve score but does it really say anything about intelligence?