Research map / Label-efficient and robust learning / Noisy labels
Rank-guided Diffusion for Noise Few-Shot Learning
Zelei Wu, Kun Zhou, Xulun Ye, Yifan Mei, Jie Hong, Jieyu Zhao
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 low-rank guidance to the diffusion refinement (ablation) | CRDProto diffusion · MiniImageNet 40% noise | accuracy: 70.84 -> 80.02 | Improved |
| add semantic-weighted prototype on top of low-rank diffusion (ablation) | CRDProto · MiniImageNet 40% noise | accuracy: 80.02 -> 81.39 | Improved |
| replace unweighted mean prototype with semantic-weighted prototype (ablation) | CRDProto · MiniImageNet 5w5s | accuracy: 78.33 -> 90.61 | Improved |
| replace unweighted mean prototype with semantic-weighted prototype, severe noise (ablation) | CRDProto · Label-swap 40% noise | accuracy: 70.22 -> 81.39 | Improved |
| add semantic-weighted prototype to reduce feature variance (ablation) | CRDProto · Ablation feature consistency | intra-class feature variance: 3.29 -> 2.25 | Improved |
| add CLIP features with diffusion refinement to ProtoNet (ablation) | ProtoNet · Label-swap 40% noise | accuracy: 57.07 -> 62.05 | Improved |
| use ViT-B/32 CLIP backbone instead of ResNet12 | CRDProto · MiniImageNet 0% label swap | accuracy: 80.25 -> 90.61 | Improved |
| use DINO ViT-S backbone instead of CLIP ViT-B/32 | CRDProto · MiniImageNet 60% label swap | accuracy: 63.77 -> 68.22 | Improved |
| use CRDProto (CLIP) instead of ECER-FSL at heavy noise | ECER-FSL · MiniImageNet 60% label swap | accuracy: 60.38 -> 63.77 | Improved |
| use CRDProto (CLIP) instead of ECER-FSL on clean data | ECER-FSL · MiniImageNet 0% label swap | accuracy: 93.26 -> 90.61 | Hurt |
| add low-rank guidance (eta > 0) to diffusion, qualitative (ablation) | CRDProto · t-SNE feature distribution | cluster compactness | Improved |
Closest papers by meaning
- Few-Shot Diffusion Models Escape the Curse of Dimensionality (NeurIPS 2024 · similarity 0.88)
- Few-shot Learner Parameterization by Diffusion Time-steps (CVPR 2024 · similarity 0.87)
- Adapting to Unknown Low-Dimensional Structures in Score-Based Diffusion Models (NeurIPS 2024 · similarity 0.87)
- Training Diffusion-based Generative Models with Limited Data (ICML 2025 · similarity 0.87)
- Noise Matters: Optimizing Matching Noise for Diffusion Classifiers (NeurIPS 2025 · similarity 0.86)
- CleanDIFT: Diffusion Features without Noise (CVPR 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.
