Research map / Image generation / Text-to-image diffusion
RAGD: Regional-Aware Diffusion Model for Text-to-Image Generation
Zhennan Chen, Yajie Li, Haofan Wang, Zhibo Chen, Zhengkai Jiang, Jun Li, Qian Wang, Jian Yang, Ying Tai
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add regional soft refinement on top of hard binding (ablation) | RAGD on Flux.1-dev · T2ICompBench | VQAScore: 0.7656 -> 0.8907 | Improved |
| add regional hard binding on top of soft refinement (ablation) | RAGD on Flux.1-dev · T2ICompBench | Spatial: 0.4939 -> 0.5193 | Improved |
| add hard binding on top of soft refinement (ablation) | RAGD on Flux.1-dev · T2ICompBench | VQAScore: 0.8413 -> 0.8907 | Improved |
| add soft refinement to hard binding (ablation) | RAGD on Flux.1-dev · T2ICompBench | Spatial: 0.2821 -> 0.5193 | Improved |
| apply RAGD to SD3.5 instead of baseline SD3.5 | SD3.5 · T2ICompBench | VQAScore: 0.8393 -> 0.8543 | Improved |
| apply RAGD to Flux instead of baseline Flux | Flux.1-dev · T2ICompBench | VQAScore: 0.7438 -> 0.8907 | Improved |
| apply RPG to SD3.5 instead of baseline SD3.5 | SD3.5 · T2ICompBench | VQAScore: 0.8393 -> 0.8340 | Hurt |
| apply RPG to SD3.5 and Flux, color attribute | SD3.5 and Flux · T2ICompBench | Color: 0.7870 -> 0.7343 (SD3.5); 0.7680 -> 0.7033 (Flux) | Hurt |
| combine RAGD with Hyper-Flux acceleration | RAGD · Inference speed | speedup: 2.5x faster | Improved |
| increase number of regions in RAGD | RAGD · Inference time | inference time | Hurt |
Later papers that compared with this method
1 later accepted paper reports RAGD as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- Layer-wise Instance Binding for Regional and Occlusion Control in Text-to-Image Diffusion Transformers (CVPR 2026 · better on 0 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- InstanceDiffusion — InstanceDiffusion: Instance-level Control for Image Generation (CVPR 2024) · better on 3 of 3 shared results
- MIGC — MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis (CVPR 2024) · better on 2 of 2 shared results
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
- R&B: Region and Boundary Aware Zero-shot Grounded Text-to-image Generation (ICLR 2024 · similarity 0.90)
- Local Conditional Controlling for Text-to-Image Diffusion Models (AAAI 2025 · similarity 0.89)
- Multi-Region Text-Driven Manipulation of Diffusion Imagery (AAAI 2024 · similarity 0.89)
- Spatial Transport Optimization by Repositioning Attention Map for Training-Free Text-to-Image Synthesis (CVPR 2025 · similarity 0.89)
- Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.89)
- LLM Blueprint: Enabling Text-to-Image Generation with Complex and Detailed Prompts (ICLR 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.
