News for hackers. Stuff that matters.

A comment on Nested Learning: A new ML paradigm for continual learning.

Re:Nested Learning: A new ML paradigm for continual learning by Bombthecat on Monday December 08, 2025 @06:47AM (#46191193)
Damn, and before that, Titan from Google: https://research.google/blog/titans-miras-helping-ai-have-lo...

We are not at the end of AI :)

Also, someone claimed that NVIDA combined diffusion and autoregression, making it 6 times faster, but couldn't find a source. Big if true!

[ Parent | on HN ]
Re:Nested Learning: A new ML paradigm for continual learning by heavymemory on Monday December 08, 2025 @06:59AM (#46191256)
Do you have a source for the NVIDIA “diffusion plus autoregression 6x faster” claim? I can’t find anything credible on that.
[ Parent | on HN ]
Re:Nested Learning: A new ML paradigm for continual learning by Bombthecat on Monday December 08, 2025 @07:10AM (#46191331)
Me neither, that's why I wrote that someone claimed that they did.

The idea is simple, in a way, with diffusion several sentences / words get predicted, but they usually are not of great quality. With auto regression they select the correct words.

Increasing quality and speed. Sounds a bit like conscious and sub-conscious to me.

[ Parent | on HN ]
Re:Nested Learning: A new ML paradigm for continual learning by Bombthecat on Monday December 08, 2025 @07:18AM (#46191397)
Ha! Found it: https://arxiv.org/abs/2511.08923

Thanks to AI search :)

[ Parent | on HN ]