SOURCE-LINKED INTELLIGENCE

Elastic Token Compression for Pixel-Space Diffusion Transformers

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Natural images concentrate their detail in a small fraction of the frame, yet diffusion models spend a full token on every patch, in every layer and at every timestep. The waste is largest in pixel-space models, with no autoencoder to absorb low-level redundancy first. Probing a pretrained pixel text-to-image transformer, we find its middle-block tokens redundant wherever the image is flat. The redundancy occupies connected, content-shaped regions, and exploiting it requires tokens with the same geometry. Cutting a Hilbert ordering of the patches provides them. Consecutive positions are always

Read original source ↗ Open in workspace

recordType

paper

region

Global