AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Diffusion Schrödinger Bridge Matching

Yuyang Shi, Valentin De Bortoli, Andrew Campbell, Arnaud Doucet

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Diffusion theory and learned samplers · 8 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 DSBM instead of DSBSchrodinger bridge solver · MNIST-EMNIST transferruntime: 30% more efficientImproved
increase noise level sigma (ablation)DSBM · CelebA 64x64 male/old to female/youngFIDMixed
use OT solver coupling (OT-CFM) instead of DSBM without OTflow-based methods · 2D tasks2-WassersteinImproved
initialize DSBM-IMF with approximate SB coupling (DSBM-IMF+) (ablation)DSBM-IMF · 2D taskssampling errorImproved
replace DSB with DSBMSB methods · 2D tasks2-WassersteinImproved
replace Rectified Flow with DSBMgenerative transport · moons-8gaussians2-WassersteinMixed
increase image resolution from 64x64 to 128x128 at sigma=1 (ablation)DSBM · CelebAalignmentImproved
replace DSB/IMF-b/RF with DSBMSB methods · 50-dim Gaussiancovariance estimatesImproved

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