AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Riemannian MeanFlow

Dongyeop Woo, Marta Skreta, Seonghyun Park, Kirill Neklyudov, Sungsoo Ahn

ICML 2026 · Image generation › Fast diffusion sampling · approach: Rectified flow straightening · 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
use x1-prediction instead of v-prediction for the flow map (ablation)Semigroup RMF · spherical helix, D=512 and 2048one-step sample qualityImproved
use xt-prediction instead of x1-prediction (ablation)Semigroup RMF · spherical helix, D=512 and 2048one-step sample qualityHurt
use semigroup objective instead of Eulerian or Lagrangian (ablation)RMF · spherical helix, D=512, v-predsample qualityImproved
add adaptive loss weighting (p=0.5) (ablation)Eulerian RMF · spherical helix, D=512sample qualityImproved
use low-frequency time embeddings (omega 0.02) instead of high-frequency (30) (ablation)RMF differential objectives · training stabilitystabilityImproved
place stop-gradient on the whole target so no higher-order derivatives are neededRMF vs GFM · training memorymemory usage: nearly 2x lowerImproved
one-step RMF instead of 100-step Fisher FMpromoter DNA design · FANTOM5 promoter, MSEMSE: 0.030 -> 0.030No clear effect
use RMF instead of FrameDiff and FrameFlow at 5 NFEprotein backbone generation · SCOPe, 5 NFEdesignability: 0.04 -> 0.82 vs FrameFlowImproved
optimize for designabilityRMF · SCOPenoveltyHurt
evaluate reward on x1 look-ahead instead of current state xt (ablation)reward-guided RMF · promoter DNA, 5 NFEMSE: 0.017 -> 0.013Improved
evaluate reward on x1 look-ahead instead of current state xt (ablation)reward-guided RMF · protein alpha-helix, 10 NFEmean alpha-helix fraction: 0.29 -> 0.45Improved
add intermediate inference noise (eta 0.25-0.45) (ablation)RMF protein sampling · SCOPescRMSD and diversityImproved

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