Research map / Label-efficient and robust learning / OOD and anomaly detection
Transfer Anomaly Detection by Inferring Latent Domain Representations
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 |
|---|---|---|---|
| add latent domain vectors (ProS vs AEAUC) | autoencoder anomaly detector · MNIST-r avg | AUC (%): 94.3(4.9) -> 96.0(3.6) | Helped |
| add latent domain vectors (ProS vs AEAUC) | autoencoder anomaly detector · Anuran Calls avg | AUC (%): 96.9(4.0) -> 96.8(3.8) | No clear effect |
| train with target normal instances (ProT vs ProS) | ProS · Anuran Calls avg | AUC (%): 96.8(3.8) -> 99.8(0.6) | Helped |
| train with target normal instances (ProT vs ProS) | ProS · Landmine avg | AUC (%): 72.4(11.) -> 72.4(11.) | No clear effect |
| add uncertainty of latent domain vector (Gaussian posterior + KL) (ablation) | ProT · Anuran Calls avg | AUC (%): 86.5(2.2) -> 99.8(0.6) | Helped |
| add uncertainty of latent domain vector (Gaussian posterior + KL) (ablation) | ProS · MNIST-r avg | AUC (%): 95.5(4.0) -> 96.0(3.6) | Helped |
| add uncertainty of latent domain vector (Gaussian posterior + KL) (ablation) | ProT · IoT avg | AUC (%): 98.1(2.6) -> 98.4(1.7) | Helped |
| add uncertainty of latent domain vector (Gaussian posterior + KL) (ablation) | ProS · Landmine avg | AUC (%): 71.1(11.) -> 72.4(11.) | Helped |
| reduce anomaly rate in source domains from 0.1 to 0.005 (ablation) | ProT · MNIST-r avg | AUC (%): 96.6(3.2) -> 87.0(8.4) | Hurt |
| use ProS instead of AE at low anomaly rate 0.005 (ablation) | anomaly detector · MNIST-r avg | AUC (%): 72.7(3.0) -> 85.0(9.8) | Helped |
| use ProT instead of AEAUC at low anomaly rate 0.005 (ablation) | anomaly detector · MNIST-r avg | AUC (%): 81.9(10.) -> 87.0(8.4) | Helped |
| use ProS inference without re-training vs ProT training | ProS · MNIST-r | inference time: 0.0027 seconds, 3918 times faster than training of ProT | Helped |
Closest papers by meaning
- A Comprehensive Augmentation Framework for Anomaly Detection (AAAI 2024 · similarity 0.86)
- Towards Multi-Domain Learning for Generalizable Video Anomaly Detection (NeurIPS 2024 · similarity 0.85)
- Beyond Normal References: Discriminative Few-Shot Anomaly Detection (NeurIPS 2026 · similarity 0.85)
- AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios (CVPR 2025 · similarity 0.85)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.85)
- Normal-Abnormal Guided Generalist Anomaly Detection (NeurIPS 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.