AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Improving Diffusion Generalization with Weak-to-Strong Segmented Guidance

Liangyu Yuan, Yufei Huang, Mingkun Lei, Tong Zhao, Ruoyu Wang, Changxi Chi, Yiwei Wang, Chi Zhang

CVPR 2026 · Image generation › Fast diffusion sampling · approach: Guidance for conditional diffusion · 2 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
segmented guidance (CFG then SLG) vs CFG alone (ablation)SD3.5-medium · MS-COCO-1KAesthetic score: 5.279 -> 5.717Improved
segmented guidance vs CFG aloneSD3.5-medium · MS-COCO-1KHPSv2.1: 29.199 -> 29.736Improved
segmented guidance vs SLG aloneSD3.5-medium · MS-COCO-1KHPSv2.1: 27.295 -> 29.736Improved
SGG vs S2 GuidanceSD3-medium · MS-COCO-1KHPSv2.1: 29.614 -> 29.541No clear effect
add weak-to-strong guidance to training target (SGG) (ablation)SiT-B/2 · ImageNet 256 conditionalFID: 31.22 -> 4.58Improved
use SGG (CFG+BR segmented) instead of MG-only training target (ablation)SiT-B/2 · ImageNet 256 conditionalFID: 5.88 -> 4.58Improved
add BR branch weak signal in training (ablation)SiT-B/2 · ImageNet 256 unconditionalFID: 61.27 -> 43.25Improved
add AG weak model in training (ablation)SiT-B/2 · ImageNet 256 unconditionalFID: 61.27 -> 45.97Improved
add SGG on top of REPA (ablation)SiT-B/2+REPA · ImageNet 256 conditionalFID: 20.46 -> 3.07Improved
move segmentation timestep tau from 0 to 0.2 (ablation)SGG training · ImageNet 256 conditionalFID: 5.88 -> 4.58Improved
move segmentation timestep tau from 0.2 to 0.4 (ablation)SGG training · ImageNet 256 conditionalFID: 4.58 -> 5.79Hurt

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