AI Paper Atlas

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

Mirror Flow Matching with Heavy-Tailed Priors for Generative Modeling on Convex Domains

Yunrui Guan, Krishna Balasubramanian, Shiqian Ma

ICLR 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Flow matching and flow sampling · 7 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
replace Gaussian prior with Student-t prior in Mirror Flow (ablation)Mirror Flow · 10-d polytope (mixture of Gaussians)MMD (x1e-3): 5.774 -> 5.552Improved
replace Gaussian prior with Student-t prior in Mirror Flow (ablation)Mirror Flow · 10-d polytope (mixture of Gaussians)KL divergence: 0.379 -> 0.339Improved
replace Gaussian prior with Student-t prior in Mirror Flow (ablation)Mirror Flow · 6-d L2 ball (mixture of Gaussians)KL divergence (x1e-2): 9.644 -> 6.944Improved
replace Gaussian prior with Student-t prior in Mirror Flow (ablation)Mirror Flow · 6-d L2 ball (mixture of Gaussians)MMD (x1e-2): 1.221 -> 1.153Improved
use Mirror t-Flow instead of Reflected Flow Matchingconstrained flow matching · 6-d L2 ball (mixture of Gaussians)KL divergence (x1e-2): 14.827 -> 6.944Improved
use Mirror t-Flow instead of Reflected Flow Matchingconstrained flow matching · 10-d polytope (mixture of Gaussians)MMD (x1e-3): 5.868 -> 5.552Improved
use Mirror t-Flow instead of Gauge Vanillaconstrained flow matching · 10-d polytope (mixture of Gaussians)feasibility: 90.653% -> 100%Improved
use Mirror Flow instead of Mirror Diffusion Model, EDM-initializedwatermarked image generation · AFHQv2 64x64 watermarked (EDM init)FID (50k): 7.29 -> 4.27Improved
use Mirror Flow instead of Mirror Diffusion Model, EDM-initializedwatermarked image generation · AFHQv2 64x64 watermarked (EDM init)CMMD: 0.170 -> 0.023Improved
use Mirror Flow instead of Mirror Diffusion Model with random initializationwatermarked image generation · AFHQv2 64x64 watermarked (random init, 24h budget)CMMD: 0.152 -> 0.177No clear effect

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