Explorative modeling: Train on the best of K guesses

(alexiglad.github.io)

23 points | by DSemba 2 hours ago ago

3 comments

  • kamranjon a few seconds ago

    This is amazing and I think will probably end up being a pretty important development.

    I was just reading this great breakdown of how diffusion Gemma works: https://newsletter.maartengrootendorst.com/p/a-visual-guide-...

    In reference to the difficulties with applying this to autoregressive LLMs - I wonder if these type of hybrids might be a good candidate for this approach.

  • FeepingCreature an hour ago

    If their numbers are real and hold up at scale, every image model trained before this is now obsolete.

  • ltsSmitty 33 minutes ago

    I don't have enough context to know if this is actually cool or not, but it seems like it!