AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Guiding Diffusion Models with Semantically Degraded Conditions

Shilong Han, Yuming Zhang, Hongxia Wang

CVPR 2026 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 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 null prompt with semantically degraded condition (CDG)SD3 Medium · MS-COCO 2017 valFID: 35.69 -> 34.05Improved
replace null prompt with CDGQwen-Image · MS-COCO 2017 valFID: 42.45 -> 39.02Improved
replace null prompt with CDGQwen-Image · MS-COCO 2017 valAesthetic Score: 2.57 -> 2.54Hurt
replace null prompt with CDGFLUX.1-dev · MS-COCO 2017 valFID: 38.55 -> 37.11Improved
replace null prompt with CDGSD3.5 · GenAI-BenchDifferentiation: 75.10 -> 78.74Improved
replace null prompt with CDGSD3 · MS-COCO 2017 valAesthetic Score: 5.66 -> 5.70 (ICG reaches 5.79)Mixed
add stratified degradation (content tokens first, then context-aggregating tokens) (ablation)CDG · MS-COCO 2017 val, SD3, Rdeg=1.1VQA Score: 86.31 -> 92.21Improved
use WPR ranking instead of random ranking inside stratified degradation (ablation)CDG · MS-COCO 2017 val, SD3, Rdeg=1.1FID: 34.17 -> 33.89No clear effect
use WPR ranking instead of reverse or random ranking in non-stratified degradation (ablation)CDG · MS-COCO 2017 val, SD3, Rdeg=1.1FID: 50.73 (reverse), 47.02 (random) -> 35.06Improved
recompute WPR mask at every denoising step instead of once (ablation)CDG · SD3, Rdeg=1.1VQA Score and time: VQA 92.21 -> 92.33, time 5.655 s -> 8.031 sNo clear effect
use one-time WPR mask computationCDG · SD3time (s): 5.456 -> 5.655 (+3.6%)Hurt
raise degradation ratio Rdeg beyond 1.0 (remove context-aggregating tokens) (ablation)CDG · SD3, CLIP Score and Aesthetic ScoreCLIP/AestheticMixed

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