AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DiFA: Inference-Time Forward-Process Alignment for Diffusion Models

Shigui Li, Delu zeng

ICML 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 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
add DiFA at NFE 8DPM-Solver++ · CIFAR-10 (32x32)FID: 8.40 -> 4.15Improved
add DiFA at NFE 12DPM-Solver++ · CIFAR-10 (32x32)FID: 3.70 -> 2.18Improved
add DiFA at NFE 20DPM-Solver++ · CIFAR-10 (32x32)FID: 2.33 -> 1.96Improved
add DiFA at NFE 25DPM-Solver++ · ImageNet-64 (class-conditional)FID: 1.83 -> 1.64Improved
add DiFA at NFE 25UniPC · ImageNet-64 (class-conditional)FID: 1.73 -> 1.63Improved
add DiFA at NFE 5Heun · ImageNet-64FID: 230.05 -> 110.20Improved
compare DiFA with EVODiff at 12 stepsDPM-Solver++ · CIFAR-10 (32x32)FID: 2.25 -> 2.18Improved
add historical consensus component (ablation)DiFA · EDM ablation (CIFAR-10)FIDImproved
add SNR gating and magnitude alignment (ablation)DiFA · EDM ablation (CIFAR-10)FIDImproved

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