Jul 2026
A Frozen Pixel-Space Diffusion Model Can Guide Itself with Its Own Samples
This work attaches a lightweight prediction head to an intermediate layer, keep the backbone frozen, and use the discrepancy between the intermediate and final predictions as a self-guidance direction during sampling to train a frozen, pretrained pixel diffusion model that can guide itself.
Zixuan Fu, Chong Wang, Lanqing Guo et al.
· arXiv.org · 2 citations