Research map / Image generation / Text-to-image diffusion
Learn, Imagine and Create: Text-to-Image Generation from Prior Knowledge
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 |
|---|---|---|---|
| replace single text encoder (AttnGAN) with two priors encoders (LeicaGAN) | AttnGAN · CUB | Inception Score: 4.36 -> 4.62 | Helped |
| replace AttnGAN with LeicaGAN | AttnGAN · Oxford-102 | Inception Score: 3.75 -> 3.92 | Helped |
| replace AttnGAN with LeicaGAN | AttnGAN · CUB* | R-precision (top-5): 81.45 -> 85.28 | Helped |
| add the text-mask encoder (shape and layout prior) (ablation) | LeicaGAN · CUB* | Inception Score: 5.60 -> 5.69 | Helped |
| add the text-mask encoder (ablation) | LeicaGAN · Oxford-102* | R-precision (top-5): 85.03 -> 85.81 | Helped |
| add local word-level attention on top of global embeddings (lambda_w 0 -> 1) (ablation) | LeicaGAN · CUB* | Inception Score: 5.63 -> 5.69 | Helped |
| add local word-level attention (lambda_w 0 -> 1) (ablation) | LeicaGAN · CUB* | R-precision (top-5): 84.10 -> 85.28 | Helped |
| add local word-level attention (lambda_w 0 -> 1) (ablation) | LeicaGAN · Oxford-102* | R-precision (top-5): 85.77 -> 85.81 | No clear effect |
| set text-image encoder weight delta_I to 0.8 vs 0.2 (ablation) | LeicaGAN · CUB* | Inception Score: 5.36 -> 5.69 | Helped |
| use only text-image encoder word features (delta_I 0.8 -> 1.0) (ablation) | LeicaGAN · CUB* | R-precision (top-5): 85.28 -> 81.11 | Hurt |
Closest papers by meaning
- Learning Visual Generative Priors without Text (CVPR 2025 · similarity 0.90)
- LLM Blueprint: Enabling Text-to-Image Generation with Complex and Detailed Prompts (ICLR 2024 · similarity 0.89)
- Text-to-Image Generation for Abstract Concepts (AAAI 2024 · similarity 0.88)
- Elucidating the design space of language models for image generation (ICML 2025 · similarity 0.88)
- ConceptBed: Evaluating Concept Learning Abilities of Text-to-Image Diffusion Models (AAAI 2024 · similarity 0.88)
- OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation (NeurIPS 2025 · similarity 0.88)
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.