EigDiff: Eigensample Diffusion - or how to significantly reduce training dataset size with marginal performance variation?
arXiv:2610.20999v1 Announce Type: new Abstract: Diffusion models are notoriously expensive to train, requiring large-scale datasets to learn how to gradually reverse the noise from images over multiple small denoising steps. Current approaches focus on distilling the number of diffusion steps, negl…
Read the full story at arXiv cs.CV ↗