AI RESEARCH
Tweedie's Formulae and Diffusion Generative Models Beyond Gaussian
arXiv CS.LG
•
ArXi:2605.19391v1 Announce Type: cross Diffusion models have achieved remarkable success in generating samples from unknown data distributions. Most popular stochastic differential equation-based diffusion models perturb the target distribution by adding Gaussian noise, transforming it into a simple prior, and then use denoising score matching, a consequence of Tweedie's formula, to learn the score function and generate clean samples from noise.