X · @teortaxesTex
· X / Twitter
RT qt cache🪷: fascinating. llms autoregressively generating discrete symbols was a victory for the neurosymbolic tradition all along. newell & simo…
RT qt cache🪷fascinating. llms autoregressively generating discrete symbols was a victory for the neurosymbolic tradition all along. newell & simon vindicated! hinton destroyed! minsky proven right again!François Chollet: Test-time training is the only form of test-time adaptation that is "pure" deep learning, as oppose