Siddharth Roheda

Pixel-space diffusion models had inherited softmax attention from sequence modeling as their token mixer, although the optimal denoiser under Gaussian noise admits a polynomial functional expansion in the observation. PoD compresses each image patch through a low-dimensional bottleneck and stacks cascaded low-rank Volterra layers that mix tokens and channels, with no activation function on the residual path. At matched parameters and FLOPs on ImageNet, PoD beats JiT in FID at every scale, with the largest gap at the G scale.

Pixel-space diffusion models had inherited softmax attention from sequence modeling as their token mixer, although the optimal denoiser under Gaussian noise admits a polynomial functional expansion in the observation. PoD compresses each image patch through a low-dimensional bottleneck and stacks cascaded low-rank Volterra layers that mix tokens and channels, with no activation function on the residual path. At matched parameters and FLOPs on ImageNet, PoD beats JiT in FID at every scale, with the largest gap at the G scale.