I believe GPT works on text and image because it has less "priors" known about
them and therefore is dumber and needs much more data to make up for the dumb
prediction however excelled at text and none other could compare to GPT-2 (?).
Maybe vision is same as text, maybe GPT simply isn't fully understand the
general theory actually!? Universal/ largest priors are the first ones you
should code / give to all brain areas.
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T48eb73fe225c230b-Mcdba93cc09c97cd706963a39
Delivery options: https://agi.topicbox.com/groups/agi/subscription
