Luca
@whitenoise
bsky.app/profile/lfoppi…
ID: 706953
26-01-2007 08:34:55
1,1K Tweet
168 Takipçi
138 Takip Edilen
Beautiful Google Research paper. LLMs can learn in context from examples in the prompt, can pick up new patterns while answering, yet their stored weights never change. That behavior looks impossible if learning always means gradient descent. The mechanisms through which this
I spoke with Dr P Arora, PhD about Marxism: One is: What’s remarkable is that Marxism has been tried. Now, of course, defenders of Marxism say it hasn’t really been tried anywhere, but certainly the people who implemented it claimed they were implementing Marxism. And this is a