cs.LGOct 29, 2026

No Channels Left Behind: Understanding Diffusion Acceleration through Residual Gates

Authors: Junha Park, Jaemyung Yu, Jaehui Hwang, Jong-Seok Lee, Dongyoon Han

Organizations: Yonsei University · NAVER AI Lab

Abstract

Training acceleration methods for diffusion transformers, such as REPA and dispersive regularization, are commonly explained by additional supervision, spatial-structure priors, or improved feature geometry. We revisit this view and uncover a striking internal change: the AdaLN-Zero residual gate αα, which directly controls how conditioned updates enter the residual stream, is substantially reorganized by these methods. This observation motivates our hypothesis that acceleration arises, at least in part, from reshaping the residual gating pathway rather than solely from the feature structures explicitly imposed by auxiliary objectives. Supporting this view, accelerated models are markedly more sensitive to perturbations of αα, and their gate geometry reflects semantic relationships among classes. We further introduce the normalized participation ratio (nPR) to characterize the channel-wise organization of the gates and find that, across REPA teachers, projector architectures, and radial-objective targets, broader gate participation accompanies better generation quality. Guided by this insight, we propose a simple teacher-free radial objective that promotes broader gate participation through feature-norm regularization, without pairwise feature comparisons. On ImageNet 256×256256\times256 with SiT-B/2, our method improves FID over dispersive loss while increasing mean gate participation over timesteps and classes, further supporting the link between gate organization and training acceleration. Our results identify residual-gate reorganization as a measurable axis of diffusion training acceleration and suggest a simple direction for further acceleration.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Where Should a Diffusion Transformer Meet Its Teacher?

    Nov 18, 2026Zhuoqi Hu, Shiwei Li, Haozhao Wang +7Diffusion TransformerRepresentation Alignment

  2. AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers

    Oct 12, 2026Javad Rajabi, Koorosh Roohi, David B. Lindell +1Diffusion TransformerDiffusion-Based Image Generation

  3. Rethinking Cross-Layer Information Routing in Diffusion Transformers

    May 20, 2026Chao Xu, Maohua Li, Qirui Li +9Diffusion TransformerResidual Learning