AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Supervised Guidance Training for Infinite-Dimensional Diffusion Models

Elizabeth Louise Baker, Alexander Denker, Jes Frellsen

ICML 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and learned samplers · 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
replace FunDPS Tweedie-approximate guidance with learned SGT guidance (k=1)FunDPS · Sparse observationsRMSE: 0.456 -> 0.373Improved
replace FunDPS guidance with learned SGT guidance (k=1)FunDPS · Sparse observationsenergy score: 3.709 -> 3.009Improved
use SGT guidance with a smaller network instead of a same-size conditional diffusionCond. Diff. (SGT size) · Sparse observationsRMSE: 0.406 -> 0.373Improved
use full-size conditional diffusion instead of SGTSGT k=1 · Sparse observationsRMSE: 0.373 -> 0.347Mixed
add smoothness preconditioner C^k, k from 0 to 1 (ablation)SGT · Sparse observationsRMSE: 0.390 -> 0.373Improved
add smoothness preconditioner C^k, k from 0 to 1 (ablation)SGT · Sparse observationsenergy score: 3.148 -> 3.009Improved
add smoothness preconditioner C^k (ablation)SGT · Heat equationRMSE: 0.0443 -> 0.0448No clear effect
use SGT instead of FunDPSFunDPS · Heat equationRMSE: 0.0519 -> 0.0443Improved
use SGT with a randomly initialised base model instead of FunDPSFunDPS · Imperfect base model, randomenergy score: 20.685 -> 6.591Improved
use SGT with a partly trained base model instead of FunDPSFunDPS · Imperfect base model, partly trainedenergy score: 3.816 -> 3.230Improved
use SGT with a fully trained base model instead of FunDPSFunDPS · Imperfect base model, fully trainedenergy score: 3.608 -> 3.190Improved
use SGT instead of conditional diffusion, 4 EFD coefficientsCond. Diff. · Shape inpainting (4 EFD coefficients)RMSE: 0.72 -> 0.63Improved

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