Research map / Label-efficient and robust learning / OOD and anomaly detection
Classification-Based Anomaly Detection for General Data
Read the paperCheck 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 GOAD instead of GEOM with Dirichlet post-processing | CIFAR-10 anomaly detection · CIFAR-10 | Average ROC-AUC (%): 86.0 -> 88.2 | Helped |
| Use GOAD instead of GEOM without Dirichlet post-processing | CIFAR-10 anomaly detection · CIFAR-10 | Average ROC-AUC (%): 81.6 -> 88.2 | Helped |
| Use GOAD instead of GEOM with Dirichlet post-processing | FashionMNIST anomaly detection · FashionMNIST | Average ROC-AUC (%): 93.5 -> 94.1 | Helped |
| Use GOAD instead of the strongest reported baseline | Arrhythmia anomaly detection · Arrhythmia | F1 score: 51.5 -> 52.0 | No clear effect |
| Use GOAD instead of FB-AE | Thyroid anomaly detection · Thyroid | F1 score: 75.0 -> 74.5 | No clear effect |
| Use GOAD instead of DAGMM | KDDCUP99 anomaly detection · KDDCUP99 | F1 score: 93.7 -> 98.4 | Helped |
| Use GOAD instead of FB-AE | KDDCUP99-Rev anomaly detection · KDDCUP99-Rev | F1 score: 95.9 -> 98.9 | Helped |
| Use open-set classification rather than closed-set softmax (ablation) | GOAD · CIFAR-10 | ROC-AUC | Helped |
| Use open-set classification rather than closed-set softmax (ablation) | GOAD · KDD | F1 score: F_1=0.99 | Hurt |
| Use random affine transformations instead of image geometric transformations (ablation) | GOAD with CNNs · CIFAR-10 | ROC-AUC | Hurt |
| Vary the margin from s=0.1 to s=1.0 (ablation) | GOAD · CIFAR-10 | Average ROC-AUC (%): 88.2 -> 88.1 | No clear effect |
| Use affine transformations rather than rotations or permutations (ablation) | GOAD · Tabular datasets | F1 score | Helped |
Closest papers by meaning
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.87)
- Normal-Abnormal Guided Generalist Anomaly Detection (NeurIPS 2025 · similarity 0.87)
- Anomaly Heterogeneity Learning for Open-set Supervised Anomaly Detection (CVPR 2024 · similarity 0.85)
- AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios (CVPR 2025 · similarity 0.85)
- Beyond Individual Input for Deep Anomaly Detection on Tabular Data (ICML 2024 · similarity 0.85)
- One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection (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.