AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Image Diffusion Preview with Consistency Solver

Fu-Yun Wang, Hao Zhou, Liangzhe Yuan, Sanghyun Woo, Boqing Gong, Bohyung Han, Ming-Hsuan Yang, Han Zhang, Yukun Zhu, Ting Liu, Long Zhao

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 1 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
train solver coefficients with RL instead of trajectory distillation (ablation)ConsistencySolver · COCO 2017 SD1.5, 5 stepsFID: 22.91 -> 20.39Improved
train solver coefficients with RL instead of trajectory distillation (ablation)ConsistencySolver · COCO 2017 SD1.5, 8 stepsFID: 19.65 -> 18.82Improved
use learned ConsistencySolver instead of Multistep DPM-SolverSD1.5 sampling · COCO 2017, 5 stepsFID: 25.87 -> 20.39Improved
use ConsistencySolver instead of EulerFLUX.1-Kontext · KontextBench, 4 stepsEdit Reward: 0.61 -> 0.73Improved
apply solver trained on SD1.5 to SDXL without retrainingConsistencySolver · COCO 2017 SDXL, 10 stepsFID: 26.32 -> 23.32Improved
use preview-then-refine workflow with 8-step ConsistencySolverDiffusion Preview · LAION promptsinference time (s): 6.35 -> 2.87Improved
use preview-then-refine workflowDiffusion Preview · LAION promptsaverage attempts: 6.00 -> 6.35Hurt
use DMD2 one-step distillation instead of the base modelDMD2 · GenEval promptsprompts satisfied within 10 attempts: 57.0% of base (w/ GAN), 47.1% (w/o GAN)Hurt
use ConsistencySolver instead of DMD2 for previewsConsistencySolver · GenEval promptsprompts satisfied within 10 attempts: 57.0% -> 94.2% of baseImproved
use solver order 4 instead of lower orders (ablation)ConsistencySolver · COCO 2017structural fidelityImproved
increase solver order from 4 to 5 (ablation)ConsistencySolver · COCO 2017consistencyNo clear effect
use depth-map reward (ablation)ConsistencySolver RL · COCO 2017consistencyImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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