Research map / Probabilistic and scientific modeling / Molecules and proteins
Few-Shot Fast-Adaptive Anomaly Detection
Ze Wang, Yipin Zhou, Rui Wang, Tsung-Yu Lin, Ashish Shah, Ser-Nam Lim
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 |
|---|---|---|---|
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Hazelnut | mIoU: 0.36 -> 0.40 | Improved |
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Cable | mIoU: 0.17 -> 0.18 | Improved |
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Cable | AUC-ROC: 0.76 -> 0.80 | Improved |
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Zipper | mIoU: 0.12 -> 0.15 | Improved |
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Leather | mIoU: 0.40 -> 0.40 | No clear effect |
| Increase sparse coding receptive field from l = 1 to l = 3 (ablation) | Ours · MVTec-AD Grid | AUC-ROC: 0.80 -> 0.81 | No clear effect |
| Add sparsity constraint (shrinkage function) in sparse coding (ablation) | Ours · MVTec-AD Leather | mIoU: 0.32 -> 0.40 | Improved |
| Add sparsity constraint (shrinkage function) in sparse coding (ablation) | Ours · MVTec-AD Hazelnut | mIoU: 0.24 -> 0.40 | Improved |
| Add sparsity constraint (shrinkage function) in sparse coding (ablation) | Ours · MVTec-AD Cable | AUC-ROC: 0.68 -> 0.80 | Improved |
| Add sparsity constraint (shrinkage function) in sparse coding (ablation) | Ours · MVTec-AD Hazelnut | reconstruction error gap abnormal vs normal: 1.7e-4 -> 2.4e-4 | Improved |
| Use smooth SigShrink instead of non-smooth RELU-like shrinkage (ablation) | Ours · training loss | convergence speed and loss | Improved |
| Replace MAML-AE with Ours (adaptive sparse coding) | few-shot anomaly detection · MVTec-AD Hazelnut | mIoU: 0.11 -> 0.40 | Improved |
Closest papers by meaning
- Beyond Normal References: Discriminative Few-Shot Anomaly Detection (NeurIPS 2026 · similarity 0.85)
- PANDA: Towards Generalist Video Anomaly Detection via Agentic AI Engineer (NeurIPS 2025 · similarity 0.85)
- Distribution Prototype Diffusion Learning for Open-set Supervised Anomaly Detection (CVPR 2025 · similarity 0.84)
- OmiAD: One-Step Adaptive Masked Diffusion Model for Multi-class Anomaly Detection via Adversarial Distillation (ICML 2025 · similarity 0.84)
- Towards Visual Discrimination and Reasoning of Real-World Physical Dynamics: Physics-Grounded Anomaly Detection (CVPR 2025 · similarity 0.84)
- One-for-More: Continual Diffusion Model for Anomaly Detection (CVPR 2025 · similarity 0.84)
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.
