Research map / Image generation / Text-to-image diffusion
Subject-driven Text-to-Image Generation via Apprenticeship Learning
Wenhu Chen, Hexiang Hu, YANDONG LI, Nataniel Ruiz, Xuhui Jia, Ming-Wei Chang, William W. Cohen
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 Delta CLIP score filtering of expert-generated data (ablation) | SuTI · DreamBench-v2 | overall human score: 0.54 -> 0.82 | Improved |
| further fine-tune SuTI on subject images for 500 steps (Dream-SuTI) (ablation) | SuTI · DreamBench-v2 | overall human score: 0.82 -> 0.87 | Improved |
| increase number of in-context demonstrations (ablation) | SuTI · DreamBench-v2 | human score | Improved |
| use apprentice SuTI instead of DreamBooth per-subject tuning (Imagen) | DreamBooth · DreamBench | DINO: 0.696 -> 0.741 | Improved |
| use SuTI instead of DreamBooth | DreamBooth · DreamBench | CLIP-T: 0.306 -> 0.304 | No clear effect |
| use SuTI instead of DreamBooth | DreamBooth · DreamBench-v2 | overall human score: 0.77 -> 0.82 | Improved |
| use SuTI instead of DreamBooth | DreamBooth · DreamBench-v2 | photorealism: 0.98 -> 0.92 | Hurt |
| use SuTI instead of Re-Imagen | Re-Imagen · DreamBench-v2 | overall human score: 0.42 -> 0.82 | Improved |
Closest papers by meaning
- Subject-driven Text-to-Image Generation via Preference-based Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
- Decoupled Textual Embeddings for Customized Image Generation (AAAI 2024 · similarity 0.87)
- Large-Scale Text-to-Image Model with Inpainting is a Zero-Shot Subject-Driven Image Generator (CVPR 2025 · similarity 0.87)
- Instruct-Imagen: Image Generation with Multi-modal Instruction (CVPR 2024 · similarity 0.87)
- Yo'Chameleon: Personalized Vision and Language Generation (CVPR 2025 · similarity 0.86)
- OneActor: Consistent Subject Generation via Cluster-Conditioned Guidance (NeurIPS 2024 · similarity 0.86)
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.
