Research map / Label-efficient and robust learning / Domain adaptation and generalization
Phasic Content Fusing Diffusion Model with Directional Distribution Consistency for Few-Shot Model Adaption
Teng Hu, Jiangning Zhang, Liang Liu, Ran Yi, Siqi Kou, Haokun Zhu, Xu Chen, Yabiao Wang, Chengjie Wang, Lizhuang Ma
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 DDC loss and ICSG on top of PCF (ablation) | phasic content fusing diffusion model · FFHQ to Cartoon, 10-shot | IS: 1.886 -> 3.410 | Improved |
| add ICSG sampling on PCF+DDC (ablation) | phasic content fusing diffusion model · FFHQ to Cartoon, 10-shot | SCS: 0.690 -> 0.837 | Improved |
| add DDC loss to PCF + ICSG (ablation) | phasic content fusing diffusion model · FFHQ to Cartoon, 10-shot | IS: 2.736 -> 3.410 | Improved |
| add PCF to DDC + ICSG (ablation) | phasic content fusing diffusion model · FFHQ to Cartoon, 10-shot | IS: 2.426 -> 3.410 | Improved |
| replace SDEdit noising with phasic content fusion (ablation) | few-shot diffusion adaptation · FFHQ to Cartoon | FID: 137.79 (SDEdit 600) -> 137.79 (PCF only), full model 119.65 | Improved |
| use DDC loss instead of IDC, RSSA or NADA losses (ablation) | phasic content fusing diffusion model · FFHQ to Sketches | FID: 146.32 (IDC) -> 47.42 | Improved |
| set phasic factor Ts to 300 rather than 0 or 800 (ablation) | phasic content fusing diffusion model · FFHQ to Cartoon | FID: 136.51 (Ts=0) and 155.93 (Ts=800) -> 119.65 | Improved |
Closest papers by meaning
- Improved Training Technique for Latent Consistency Models (ICLR 2025 · similarity 0.86)
- Improved Techniques for Training Consistency Models (ICLR 2024 · similarity 0.85)
- Compositional Discrete Latent Code for High Fidelity, Productive Diffusion Models (NeurIPS 2025 · similarity 0.85)
- APT: Adaptive Personalized Training for Diffusion Models with Limited Data (CVPR 2025 · similarity 0.85)
- Zero-Shot Adaptation of Parameter-Efficient Fine-Tuning in Diffusion Models (ICML 2025 · similarity 0.85)
- Boosting Alignment for Post-Unlearning Text-to-Image Generative Models (NeurIPS 2024 · similarity 0.85)
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.
