Research map / Image generation / Editing and style transfer
A Training-Free Style-Personalization via SVD-Based Feature Decomposition
Kyoungmin Lee, Jihun Park, Jongmin Gim, Wonhyeok Choi, Kyumin Hwang, Jaeyeul Kim, Sunghoon Im
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 |
|---|---|---|---|
| replace dominant style feature directly (REP) instead of blending the SVD principal component (ablation) | Infinity-2B · Parti-190 | S_txt: 0.348 -> 0.279 | Hurt |
| replace dominant style feature directly (REP) (ablation) | Infinity-2B · Parti-190 | S_img: 0.559 -> 0.696 | Improved |
| add Principal Feature Blending (SVD-based) (ablation) | Infinity-2B · Parti-190 | S_img: 0.559 -> 0.631 | Improved |
| add Principal Feature Blending (SVD-based) (ablation) | Infinity-2B · Parti-190 | S_txt: 0.348 -> 0.321 | Hurt |
| use PFB instead of direct replacement (ablation) | Infinity-2B · Parti-190 | S_txt: 0.279 -> 0.321 | Improved |
| add Structural Attention Correction on top of PFB (ablation) | Infinity-2B · Parti-190 | S_txt: 0.321 -> 0.334 | Improved |
| add Structural Attention Correction on top of PFB (ablation) | Infinity-2B · Parti-190 | S_img: 0.631 -> 0.630 | No clear effect |
| add SAC to PFB (ablation) | Infinity-2B · Parti-190 | Harmonic score: 0.426 -> 0.437 | Improved |
| compare to training-free IP-Adapter and StyleAligned | Ours · Parti-190 | S_txt: 0.302/0.315 -> 0.334 | Improved |
| compare inference time to tuning-based methods | Ours · Parti-190 | Inference time (s): 3.58 vs 10.13 - 698.98 | Improved |
| compare to StyleAligned on harmonic score | Ours · Parti-190 | Harmonic score: 0.437 vs 0.438 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CSGO — CSGO: Content-Style Composition in Text-to-Image Generation (NeurIPS 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
- CRAFT-LoRA: Content-Style Personalization via Rank-Constrained Adaptation and Training-Free Fusion (CVPR 2026 · similarity 0.89)
- FineStyle: Fine-grained Controllable Style Personalization for Text-to-image Models (NeurIPS 2024 · similarity 0.87)
- Conditional Balance: Improving Multi-Conditioning Trade-Offs in Image Generation (CVPR 2025 · similarity 0.87)
- CLeaR: A Unified Framework for Resolving the Leakage–Degradation Dilemma in Style Transfer (NeurIPS 2026 · similarity 0.86)
- SigStyle: Signature Style Transfer via Personalized Text-to-Image Models (AAAI 2025 · similarity 0.86)
- CSGO: Content-Style Composition in Text-to-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.
