AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

LeapAlign: Post-training Flow Matching Models at Any Generation Step by Building Two-Step Trajectories

Zhanhao Liang, Tao Yang, Jie Wu, Chengjian Feng, Liang Zheng

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Guidance and alignment for flow matching · 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
use gradient discounting alpha=0.3 instead of removing the nested gradient (alpha=0) (ablation)LeapAlign · HPDv2 testHPSv2.1: 0.4064 -> 0.4092Improved
use gradient discounting alpha=0.3 instead of no discounting (alpha=1) (ablation)LeapAlign · HPDv2 testHPSv2.1Improved
use leap trajectory design even without nested gradient, versus DRTuneLeapAlign (alpha=0) · HPDv2 testHPSv2.1: 0.3882 -> 0.4064Improved
add trajectory-similarity weighting at both x_j and x_0 (ablation)LeapAlign · HPDv2 testHPSv2.1Improved
use two-step instead of one-step leap trajectory (ablation)LeapAlign · HPDv2 testHPSv2.1Improved
use three-step instead of two-step leap trajectory (ablation)LeapAlign · HPDv2 testHPSv2.1No clear effect
train on timestep range [0,1] instead of [0,1/2] (ablation)LeapAlign · GenEvalGenEval scoreImproved
select k and j randomly instead of fixed distance 1/2 (ablation)LeapAlign · HPDv2 testHPSv2.1Improved
evaluate reward on generated x_0 instead of one-step prediction x_0|j (ablation)LeapAlign · HPDv2 testHPSv2.1Improved
raise loss threshold lambda from 0.55 to 0.95 (ablation)LeapAlign · HPDv2 testHPSv2.1: 0.4092 -> 0.4023Hurt
lower loss threshold lambda from 0.55 to 0.35 (ablation)LeapAlign · HPDv2 testHPSv2.1: 0.4092 -> 0.3860Hurt
replace DRTune with LeapAlignFlux · GenEvalGenEval overall: 0.7101 -> 0.7420Improved

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