Research map / Image generation / Text-to-image diffusion
Unified Customized Generation by Disentangled Reward Modeling
Shaojin Wu, Mengqi Huang, Yufeng Cheng, Wenxu Wu, Jiahe Tian, Yiming Luo, Fei Ding, Qian He
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 auxiliary style reward (ablation) | USO · USO-Bench style-subject-driven | CSD: 0.413 -> 0.492 | Improved |
| add auxiliary style reward (ablation) | USO · USO-Bench subject-driven | CLIP-I: 0.619 -> 0.647 | Improved |
| add auxiliary style reward (ablation) | USO · USO-Bench style-driven | CSD: 0.491 -> 0.556 | Improved |
| use disentangled encoders (SigLIP for style, VAE for content) instead of one shared VAE (ablation) | USO · USO-Bench style-subject-driven | CSD: 0.382 -> 0.492 | Improved |
| use disentangled encoders instead of one shared VAE (ablation) | USO · USO-Bench subject-driven | CLIP-I: 0.594 -> 0.647 | Improved |
| train OmniStyle on the layout-shifted USO dataset (ablation) | OmniStyle · USO-Bench style-subject-driven | CLIP-T: 0.229 -> 0.277 | Improved |
| train UNO on the USO dataset (ablation) | UNO · USO-Bench subject-driven | CLIP-I / CLIP-T: CLIP-I 0.605 -> 0.596; CLIP-T 0.264 -> 0.278 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 on most shared results.
- OmniStyle — OmniStyle: Filtering High Quality Style Transfer Data at Scale (CVPR 2025) · better on 2 of 2 shared results
- StyleStudio — StyleStudio: Text-Driven Style Transfer with Selective Control of Style Elements (CVPR 2025) · better on 1 of 1 shared result
- CSGO — CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 2025) · better on 1 of 1 shared result
- RealCustom++ — RealCustom: Narrowing Real Text Word for Real-Time Open-Domain Text-to-Image Customization (CVPR 2024) · better on 0 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
- Direct Consistency Optimization for Robust Customization of Text-to-Image Diffusion models (NeurIPS 2024 · similarity 0.87)
- AlignedGen: Aligning Style Across Generated Images (NeurIPS 2025 · similarity 0.87)
- Conditional Balance: Improving Multi-Conditioning Trade-Offs in Image Generation (CVPR 2025 · similarity 0.86)
- DisEnvisioner: Disentangled and Enriched Visual Prompt for Customized Image Generation (ICLR 2025 · similarity 0.86)
- CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 2025 · similarity 0.86)
- OneIG-Bench: Omni-dimensional Nuanced Evaluation for Image Generation (NeurIPS 2025 · 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.
