This media is not supported in your browser
VIEW IN TELEGRAM
Mercury, the first commercial-grade diffusion large language model (dLLM)! This's so interesting.
Most of the LLMs you've been seeing are clones as far as the core modeling approach goes. They're all trained autoregressively", i.e. predicting tokens from left to right.
Diffusion is different - it doesn't go left to right, but all at once. You start with noise and gradually denoise into a token stream.
Most of the image / video generation AI tools actually work this way and use Diffusion, not Autoregression.
#Mercury #AIML
@Dagmawi_Babi
Most of the LLMs you've been seeing are clones as far as the core modeling approach goes. They're all trained autoregressively", i.e. predicting tokens from left to right.
Diffusion is different - it doesn't go left to right, but all at once. You start with noise and gradually denoise into a token stream.
Most of the image / video generation AI tools actually work this way and use Diffusion, not Autoregression.
#Mercury #AIML
@Dagmawi_Babi