upvote
Thank you! If were able to natively update weights as a way of learning in real time during inference, that would be one thing. But using scripts to modify weights in the same way as human researchers (and in all likelihood trained on a massive corpus of them doing so) is just looped code generation which it's been doing for years now. It can be impressive / interesting, but it's not a phase shift.
reply