masteranza

4 माह पहलेdiscuss

We can surely fix it and we probably should. However, I don't think AI is doing any worse here than friends advice when they here a one sided story. The only difference being that it's not getting studied.

Conversely, AI chatbots are great mediators if both parties are present in the conversation.

masteranza

4 माह पहलेdiscuss

A few years ago I've made this simple thought experiment to convince myself that LLM's won't achieve superhuman level (in the sense of being better than all human experts):

Imagine that we made an LLM out of all dolphin songs ever recorded, would such LLM ever reach human level intelligence? Obviously and intuitively the answer is NO.

Your comment actually extended this observation for me sparking hope that systems consuming natural world as input might actually avoid this trap, but then I realized that tool use & learning can in fact be all that's needed for singularity while consuming raw data streams most of the time might actually be counterproductive.

masteranza

7 माह पहलेdiscuss

It's basically way better than LoRA under all respects and could even be used to speed up inference. I wonder whether the big models are not using it already... If not we'll see a blow up in capabilities very, very soon. What they've shown is that you can find the subset of parameters responsible for transfer of capability to new tasks. Does it apply to completely novel tasks? No, that would be magic. Tasks that need new features or representations break the method, but if it fits in the same domain then the answer is "YES".

Here's a very cool analogy from GPT 5.1 which hits the nail in the head in explaining the role of subspace in learning new tasks by analogy with 3d graphics.

  Think of 3D character animation rigs:
  
   • The mesh has millions of vertices (11M weights).
  
   • Expressions are controlled via:
  
   • “smile”
  
   • “frown”
  
   • “blink”
  
  Each expression is just:
  
  mesh += α_i \* basis_expression_i
  
  Hundreds of coefficients modify millions of coordinates.

masteranza

2 वर्ष पहलेdiscuss

I hope I’m wrong, as I didn’t even know who Pavel Durov was until now, but the first thought that came to mind was that it’s a show of power to intimidate Elon Musk.

masteranza

2 वर्ष पहलेdiscuss

Looks interesting, but the contrast between white background and actual windows is preventing me to use it. How can it be changed?

comments