AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DyDiT++: Diffusion Transformers With Timestep and Spatial Dynamics for Efficient Visual Generation

Wangbo Zhao, Yizeng Han, Jiasheng Tang, Kai Wang, Hao Luo, Yibing Song, Gao Ming Huang, Fan Wang, Yang You

TPAMI 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Spatial-wise Dynamic Token (SDT) to TDW (ablation)DyDiT-S lambda=0.5 · ImageNetFID: 31.89 -> 28.75Improved
add Timestep-wise Dynamic Width (TDW) to SDT (ablation)DyDiT-S lambda=0.5 · ImageNetFID: 70.06 -> 28.75Improved
add SDT to TDW (ablation)DyDiT-S lambda=0.5 · Fine-grained avgFID: 20.93 -> 16.94Improved
use learnable router instead of random (ablation)TDW · ImageNetFID: 124.38 -> 28.75Improved
use token-level bypass instead of layer skipping (ablation)SDT · ImageNetFID: 30.95 -> 28.75Improved
use token-level bypass instead of layer skipping (ablation)SDT · Fine-grained avgFID: 18.29 -> 16.94Improved
use learned dynamic width instead of manual (ablation)TDW · Fine-grained avgFID: 28.10 -> 20.93Improved
dynamic architecture lambda=0.5 vs static DiT-XLDiT-XL · ImageNet 256FLOPs / FID: 118 -> 57.88 FLOPs, 2.27 -> 2.07 FIDImproved
dynamic architecture vs pruning (magnitude) on DiT-SDiT-S · Fine-grained avgFID: 35.44 -> 13.98Improved
dynamic architecture lambda=0.5 vs original DiT-SDiT-S · Fine-grained avgFID: 12.27 -> 13.98Hurt
combine DyDiT with DeepCache interval 5DiT-XL · ImageNet 256FID: 6.73 -> 3.37Improved
combine DyDiT with 10-step DPM-solver++DiT-XL · ImageNet 256FID: 11.66 -> 11.10Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy