Research map / Image generation / Text-to-image diffusion
Dynamic Image Prompt Adapter for Scalable Zero-shot Personalized Text-to-Image Generation
Zhizhong Wang, Tianyi Chu, Richard Wang, Nanyang Wang, Kehan Li
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 Dynamic Decoupling Strategy (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.627 -> 0.650 | Improved |
| add Dynamic Decoupling Strategy (ablation) | DynaIP · DynaIP-Bench (multi-subject) | CP·PF: 0.272 -> 0.615 | Improved |
| add Dynamic Decoupling Strategy (ablation) | DynaIP · DynaIP-Bench (multi-subject) | PF: 0.545 -> 0.997 | Improved |
| add Dynamic Decoupling Strategy (ablation) | DynaIP · DreamBench++ (single-subject) | PF: 0.799 -> 0.934 | Improved |
| add Dynamic Decoupling Strategy (ablation) | DynaIP · DreamBench++ (single-subject) | CP: 0.785 -> 0.696 | Hurt |
| use HMoE routing instead of add fusion (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.633 -> 0.650 | Improved |
| use HMoE routing instead of concat fusion (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.629 -> 0.650 | Improved |
| use HMoE routing instead of add fusion (ablation) | DynaIP · DynaIP-Bench (multi-subject) | CP·PF: 0.606 -> 0.615 | Improved |
| use HMoE routing instead of concat fusion (ablation) | DynaIP · DynaIP-Bench (multi-subject) | CP·PF: 0.602 -> 0.615 | Improved |
| fuse shallow, mid and deep CLIP experts instead of only shallow (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.579 -> 0.650 | Improved |
| fuse shallow, mid and deep CLIP experts instead of only mid (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.622 -> 0.650 | Improved |
| fuse shallow, mid and deep CLIP experts instead of only deep (ablation) | DynaIP · DreamBench++ (single-subject) | CP·PF: 0.456 -> 0.650 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- XVerse — XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation (NeurIPS 2025) · better on 2 of 2 shared results
- MS-Diffusion — MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance (ICLR 2025) · better on 2 of 2 shared results
- DisEnvisioner — DisEnvisioner: Disentangled and Enriched Visual Prompt for Customized Image Generation (ICLR 2025) · better on 1 of 1 shared result
- OminiControl — OminiControl: Minimal and Universal Control for Diffusion Transformer (ICCV 2025) · better on 1 of 1 shared result
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
- Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models (CVPR 2024 · similarity 0.87)
- Devil is in the Detail: Towards Injecting Fine Details of Image Prompt in Image Generation via Conflict-free Guidance and Stratified Attention (CVPR 2025 · similarity 0.87)
- MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance (ICLR 2025 · similarity 0.87)
- Imagine Before You Draw: Visual Prompt Engineering for Image Generation (NeurIPS 2026 · similarity 0.87)
- A User-Friendly Framework for Generating Model-Preferred Prompts in Text-to-Image Synthesis (AAAI 2024 · similarity 0.87)
- Textualize Visual Prompt for Image Editing via Diffusion Bridge (AAAI 2025 · similarity 0.87)
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.
