Research map / Image generation / Text-to-image diffusion
Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| scale frozen text encoder size (T5 small to XXL) (ablation) | Imagen · MS-COCO / DrawBench | FID and CLIP score | Helped |
| scale U-Net size 300M to 2B (ablation) | Imagen base 64x64 · MS-COCO | CLIP-FID tradeoff | Helped |
| use T5-XXL instead of CLIP text encoder (ablation) | Imagen · MS-COCO | FID/CLIP | No clear effect |
| use T5-XXL instead of CLIP text encoder (ablation) | Imagen · DrawBench | human preference | Helped |
| add noise conditioning augmentation to super-resolution models (ablation) | Imagen 64->256 SR · MS-COCO | CLIP and FID | Helped |
| cross attention over text embeddings instead of mean/attention pooling (ablation) | Imagen base 64x64 · MS-COCO | CLIP-FID tradeoff | Helped |
| dynamic thresholding instead of static thresholding (ablation) | Imagen base 64x64 · MS-COCO | CLIP score | Helped |
| increase inference aug_level noise for SR (ablation) | Imagen 64->256 SR · MS-COCO | FID-10K | Hurt |
Closest papers by meaning
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.91)
- Exploring the Deep Fusion of Large Language Models and Diffusion Transformers for Text-to-Image Synthesis (CVPR 2025 · similarity 0.91)
- Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models (ICLR 2024 · similarity 0.91)
- Dual Diffusion for Unified Image Generation and Understanding (CVPR 2025 · similarity 0.89)
- Image Regeneration: Evaluating Text-to-Image Model via Generating Identical Image with Multimodal Large Language Models (AAAI 2025 · similarity 0.89)
- De-Diffusion Makes Text a Strong Cross-Modal Interface (CVPR 2024 · similarity 0.89)
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.