AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Accelerated Likelihood Maximization for Diffusion-based Versatile Content Generation

Hyunsoo Lee, Inwoo Hwang, Young Min Kim

ECCV 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and design space · 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
add the likelihood maximization over the unobserved region (ablation)ALM · AFHQ inpaintingM-SSIM: 0.300 -> 0.351Improved
add the likelihood maximization over the unobserved region (ablation)ALM · CelebA-HQ inpaintingM-SSIM: 0.332 -> 0.417Improved
add the conditional likelihood term (ablation)ALM · AFHQ inpaintingM-SSIM: 0.323 -> 0.351Improved
add the conditional likelihood term (ablation)ALM · CelebA-HQ inpaintingM-SSIM: 0.357 -> 0.417Improved
add the joint log-density term (ablation)ALM · AFHQ inpaintingM-SSIM: 0.327 -> 0.351Improved
add the joint log-density term (ablation)ALM · CelebA-HQ inpaintingM-SSIM: 0.385 -> 0.417Improved
use one-step approximation instead of iterative updates (ablation)ALM · AFHQ inpaintingM-SSIM: 0.298 -> 0.351No clear effect
use one-step approximation instead of iterative updates (ablation)ALM · CelebA-HQ inpaintingLPIPS: 0.277 -> 0.251No clear effect
ALM vs SyncSDE on inpaintingALM · AFHQ inpaintingM-SSIM: 0.302 -> 0.351Improved
ALM vs SDI on AFHQ MSEALM · AFHQ inpaintingMSE: 0.140 -> 0.143Hurt
ALM vs SyncSDE on wide image generationALM · Wide image generationKID: 51.84 -> 42.98Improved
swap backbone from SD to FLUXALM · AFHQ inpaintingM-SSIM: 0.351 -> 0.277Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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