Regular LLMs emit tokens sequentially for humans, but what if we created a Diffusion-based LLM?
We'd still read it sequentially
Please Login to reply.
No replies yet.
