AI Paper Atlas

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

Skews in the Phenomenon Space Hinder Generalization in Text-to-Image Generation

Yingshan Chang, Yasi Zhang, Zhiyuan Fang, Ying Nian Wu, Yonatan Bisk, Feng Gao

ECCV 2024 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 4 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
make training set visually complete and balanced (ablation)pixel diffusion on What'sUp · What'sUp subsetB (visual skew)Final accuracy: 18.75 -> 48.64Improved
increase visual balance while keeping full visual completeness (ablation)pixel diffusion on What'sUp · What'sUp subsetB (visual skew)Final accuracy: 10.17 -> 48.64Improved
make training set linguistically complete and balanced (ablation)pixel diffusion on What'sUp · What'sUp subsetB (linguistic skew)Final accuracy: 57.14 -> 67.29Improved
visual skew versus equal linguistic skew (ablation)pixel diffusion on What'sUp · What'sUp subsetBFinal accuracy: visual skew worse: 10.17 vs 62.71 at BLC 73Hurt
increase training coverage (vanilla) without improving completeness or balance (ablation)synthetic icon diffusion · synthetic iconstesting accuracyNo clear effect
increase number of concepts N (ablation)synthetic icon diffusion · synthetic iconstesting accuracyImproved
improve linguistic completeness and balance (ablation)synthetic icon diffusion · synthetic icons, final accuracyfinal testing accuracyNo clear effect
use T5-small text encoder instead of CLIP-B/16 (ablation)text-to-image diffusion · synthetic iconstesting accuracyImproved
add image positional embeddings (ablation)pixel diffusion UNet · synthetic icons, N=30testing accuracyImproved
use T5-flan-xxl instead of CLIP for spatial role probing (ablation)text encoder probe · role probing out-of-distributionout-of-distribution accuracy: 74.5 -> 99.8Improved
move from pixel to latent diffusion (ablation)text-to-image diffusion · synthetic and natural imagesgeneralization trendsNo clear effect

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