X · @teortaxesTex
· X / Twitter
This is really impressive on performance, but such «iterative refinement» use case seems almost maximally unconvincing. I have been a diffusion LM b…
This is really impressive on performance, but such «iterative refinement» use case seems almost maximally unconvincing.I have been a diffusion LM bear for years, maybe Google will finally prove me wrong by scaling this up.Brendan O'Donoghue: Text diffusion isn't just for batch-size = 1, it has latency and throughput be