News for hackers. Stuff that matters.
Re:DiffusionGemma Technical Report by yorwba on Thursday August 20, 2026 @02:21PM (#49378260)
Diffusion language models work with a discrete output space, unlike image models that repeatedly refine a continuous output, so they don't do the noise-prediction thing anyway.
Re:DiffusionGemma Technical Report by LarsDu88 on Thursday August 20, 2026 @02:54PM (#49378620)
Ok, you are correct. These models don't train noise predictors at all unlike first gen image diffusion