AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture

Hui Li, Jiayue Lyu, Fu-Yun Wang, Kaihui Cheng, Siyu Zhu, Jingdong Wang

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 0 citations (OpenAlex)

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
post-train with MixFlow slowed interpolation mixtureSiT-B/2 · ImageNet 256x256 (w/o guidance)gFID: 17.97 -> 15.64Improved
post-train with MixFlowSiT-XL/2 · ImageNet 256x256 (w/ guidance)gFID: 2.15 -> 1.99Improved
add MixFlow post-trainingRAE-XL · ImageNet 256x256gFID: 1.51 -> 1.43 (w/o guidance); 1.13 -> 1.10 (w/ guidance)Improved
add MixFlow post-trainingREPA-XL/2 · ImageNet 256x256gFID: 6.90 -> 6.28 (w/o guidance); 1.65 -> 1.59 (w/ guidance)Improved
sample slowed timestep mt from U[0,t] instead of U[0,1] (ablation)MixFlow SiT-B · ImageNet 256x256 (w/o guidance)gFID: 18.27 -> 16.57Improved
sample t from Beta(2,1) instead of U[0,1] (ablation)MixFlow SiT-B · ImageNet 256x256gFID: 16.57 -> 15.64 (w/o guidance); 4.25 -> 3.93 (w/ guidance)Improved
standard post-training for same extra steps instead of MixFlow (ablation)SiT-B · ImageNet 256x256gFID: 17.97 -> 17.96 (w/o guidance); 4.46 -> 4.46 (w/ guidance)No clear effect
vary mixture range gamma from 0.7 to 1.0 (ablation)MixFlow SiT-B · ImageNet 256x256gFID: 15.65, 15.64, 15.64, 15.64 (w/o guidance)No clear effect
use gamma=0 (no mixture) (ablation)MixFlow SiT-B · ImageNet 256x256gFID: 17.97 -> 18.00 (w/o guidance)Hurt
MixFlow vs Time-Shift samplerSiT-B/2 · ImageNet 256x256 (w/o guidance)gFID: 17.97 -> 18.12 (Time-Shift)Hurt
fewer sampling steps (250 to 20) (ablation)MixFlow SiT-B/2 · ImageNet 256x256 (w/o guidance)gFID gain: gain 2.33 -> 2.58Improved
add MixFlow post-trainingSD 3.5-medium fine-tuned · GenEval/DPG-Bench/T2I-CompBenchAvg scores: DPG-Bench 84.87 -> 87.49 (40 steps)Improved

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