AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Flow Along the $K$-Amplitude for Generative Modeling

weitao Du, Jiasheng Tang, Shuning Chang, Yu Rong, Fan Wang, Shengchao Liu

ICLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Flow matching and flow sampling · 3 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
flow along K-amplitude (wavelet) instead of latent flow matchingLFM DiT L/2 · CelebA-HQ 256 unconditionalFID: 5.28 -> 4.99Improved
flow along K-amplitude (Fourier) instead of latent flow matchingLFM DiT L/2 · CelebA-HQ 256 unconditionalRecall: 0.48 -> 0.47No clear effect
use three scaling components instead of two (ablation)K-Flow wavelet · LSUN Church 256 unconditionalFID: 5.37 -> 5.19Improved
K-Flow vs latent diffusion modelLDM · LSUN Church 256 unconditionalFID: 4.02 -> 5.19Hurt
use K-Flow Fourier with cfg=1.5LFM DiT L/2 · ImageNet class-conditionalFID: 2.78 -> 2.77No clear effect
K-Flow wavelet without guidanceLFM DiT L/2 · ImageNet class-conditionalFID: 14.0 -> 17.8Hurt
drop class label in high-scaling segments of generation (ablation)K-Flow Fourier vs LFM · ImageNet class-conditionalclass-dropping ratio: 3.25 -> 1.49Improved
flow along K-amplitude for molecular assemblyAssembleFlow · COD-Cluster17-AllPM (atom): 7.37 -> 7.23Improved

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