2 ms·
The problem with this approach to text generation is that it's still not flexible enough. If during inference the model changes its mind and wants to output som
by rafaelero 11mo ago
The problem with this approach to text generation is that it's still not flexible enough. If during inference the model changes its mind and wants to output something considerably different it can't because there are too many tokens already in place.
- didibus 11mo agoCould maybe be solved by reintroducing noise steps in between denoising step?
- nodja 11mo agoThat's not true, you could just have looked at the first gif animation in the OP and seen that tokens disappear, the only part that stays untouched is the prompt, adding noise is part of the diffusion process and the code that does it is even posted in the article (ctrl+f "def diffusion_collator").
- rafaelero 11mo agoLooks like you are correct.
- oezi 11mo agoDidn't anybody add backspace to an LLM's output token set yet?