• 1 Post
  • 76 Comments
Joined 1 year ago
cake
Cake day: June 4th, 2025

help-circle















  • You still lose the internal state between each token in the database output. It would let it plan, but it would still be externalizing that planning, one token at a time. Condensing all of the internal state into a single token at a time still means huge losses in detail as well as fragmentation of responses, resulting in all the problems that you see with LLMs.

    Somehow the actual internal state needs to not only be preserved, but fed back into itself. That’s how brains work. Condensing it into tokens isn’t enough.




  • SparroHawc@lemmy.ziptoGaming@lemmy.worldMMMMMM SLOOOP
    link
    fedilink
    English
    arrow-up
    7
    ·
    4 months ago

    In order for it to NOT be shit, it’s going to need to be trained on different visual styles - which will utterly balloon the model size. It already takes a whole second GPU to run. Otherwise, everything that uses it will slide towards a single, generic pretty-people style.

    If you WANT your game to look like airbrushed plastic surgery obsessed models with flawless makeup, then go hog, I guess - but I’d rather different games have different looks, myself.