Research map / Detection and segmentation / Open-set and few-shot detection
Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use student-teacher regression on pretrained discriminative embeddings for anomaly segmentation | MVTec AD · MVTec AD | normalized area under PRO curve: 0.790 -> 0.857 | Helped |
| Combine anomaly scores from receptive fields p=17, 33, and 65 (ablation) | Student-teacher · MVTec AD | normalized area under PRO curve: 0.857 -> 0.914 | Helped |
| Add descriptor compactness loss Lc to knowledge-distillation loss Lk (ablation) | Student-teacher teacher embeddings · CIFAR-10 | ROC-AUC: 0.8021 -> 0.8196 | Helped |
| Add descriptor compactness loss Lc to knowledge-distillation loss Lk (ablation) | Student-teacher teacher embeddings · MNIST | ROC-AUC: 0.9917 -> 0.9935 | Helped |
| Add triplet metric-learning loss Lm to Lk + Lc (ablation) | Student-teacher teacher embeddings · CIFAR-10 | ROC-AUC: 0.8196 -> 0.8035 | Hurt |
| Add triplet metric-learning loss Lm to Lk + Lc (ablation) | Student-teacher teacher embeddings · MNIST | ROC-AUC: 0.9935 -> 0.9926 | Hurt |
| Replace knowledge-distillation loss Lk with triplet loss Lm while retaining Lc (ablation) | Student-teacher teacher embeddings · CIFAR-10 | ROC-AUC: 0.8196 -> 0.7940 | Hurt |
| Replace knowledge-distillation loss Lk with triplet loss Lm while retaining Lc (ablation) | Student-teacher teacher embeddings · MNIST | ROC-AUC: 0.9935 -> 0.9935 | No clear effect |
| Use larger receptive fields (ablation) | Student-teacher · MVTec AD categories | normalized area under PRO curve: 0.795 -> 0.893 (Carpet); 0.943 -> 0.725 (Wood) | Mixed |
Closest papers by meaning
- Unlocking the Potential of Reverse Distillation for Anomaly Detection (AAAI 2025 · similarity 0.88)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.86)
- ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining (NeurIPS 2025 · similarity 0.86)
- Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection (CVPR 2025 · similarity 0.86)
- CNC: Cross-modal Normality Constraint for Unsupervised Multi-class Anomaly Detection (AAAI 2025 · similarity 0.85)
- A Dataset for Semantic Segmentation in the Presence of Unknowns (CVPR 2025 · 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.