AI Paper Atlas

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

ERNIE-ViLG 2.0: Improving Text-to-Image Diffusion Model With Knowledge-Enhanced Mixture-of-Denoising-Experts

Zhida Feng, Zhenyu Zhang, Xintong Yu, Yewei Fang, Lanxin Li, Xuyi Chen, Yuxiang Lu, Jiaxiang Liu, Weichong Yin, Shikun Feng, Yu Sun, Li Chen and 3 more

CVPR 2023 · Image generation › Text-to-image diffusion · approach: Compositional text-to-image guidance · 93 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 textual and visual knowledge enhancement (ablation)ERNIE-ViLG 2.0 · text-to-image generationimage fidelity, image-text alignment, and convergence speedImproved
Use object knowledge without synthetic descriptions (ablation)ERNIE-ViLG 2.0 · text-to-image generationoverall performanceNo clear effect
Add synthetic descriptions alongside object knowledge (ablation)ERNIE-ViLG 2.0 · text-to-image generationusefulness of visual knowledgeImproved
Add textual knowledge (ablation)ERNIE-ViLG 2.0 · fine-grained semantic controlCLIP Score: +0.0051 CLIP Score on Counterfactual; +0.0041 on ColorImproved
Add visual knowledge (ablation)ERNIE-ViLG 2.0 · fine-grained image-text alignmentCLIP Score: +0.0080 CLIP Score on Counterfactual; +0.0047 on CountingImproved
Combine textual and visual knowledge (ablation)ERNIE-ViLG 2.0 · ViLG-300 ComplexCLIP Score: +0.0074 CLIP ScoreImproved
Increase denoising experts from 1 to 10 (ablation)ERNIE-ViLG 2.0 · MS-COCO 256×256 zero-shot, without rerankingFID-30k: 8.07 -> 7.23Improved
Use multiple timestep-specific experts rather than one expert, at matched training samples (ablation)ERNIE-ViLG 2.0 · text-to-image generationoverall performanceImproved
Increase the number of denoising experts (ablation)ERNIE-ViLG 2.0 · text-to-image generationimage qualityImproved
Use 10 denoising experts instead of one (ablation)ERNIE-ViLG 2.0 · MS-COCO 256×256 zero-shotFID-30k: 8.07 -> 7.23 without reranking; 7.62 -> 6.75 with rerankingImproved

Later papers that compared with this method

2 later accepted papers report ERNIE-ViLG 2.0 as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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