Research map / Time series / Anomaly detection
DoKnowAD: Calibrating Normal Representations with Refined Domain Knowledge to Enhance Time Series Anomaly Detection
Shiwang Xing, Jianwei Niu, Tao Ren
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 refined AuxSets (ablation) | DoKnowAD · ECG (UCR) | F1-A (%): 56.12 -> 82.49 | Improved |
| refine AuxSets by hypervolume discrepancy instead of using all (ablation) | DoKnowAD · ECG (UCR) | F1-A (%): 64.49 -> 90.15 | Improved |
| refine AuxSets by hypervolume discrepancy instead of using all (ablation) | DoKnowAD · EPG (UCR) | F1-A (%): 66.07 -> 82.46 | Improved |
| add contrastive learning (ablation) | DoKnowAD · ECG (UCR) | F1-A (%): 82.49 -> 90.15 | Improved |
| add contrastive learning (ablation) | DoKnowAD · SMD | F1-A (%): 94.62 -> 95.41 | Improved |
| add prototype-guided scoring (ablation) | DoKnowAD · ECG (UCR) | F1-A (%): 69.49 -> 90.15 | Improved |
| add prototype-guided scoring (ablation) | DoKnowAD · EPG (UCR) | F1-A (%): 79.66 -> 82.46 | Improved |
| add all unrefined AuxSets to training (ablation) | DoKnowAD · ECG (UCR) | F1-A (%): 56.12 -> 64.49 | Improved |
| add all unrefined AuxSets to training (ablation) | baseline TSAD models · SMD | F1-A (%): up to 3% drop for baselines | Hurt |
| select top-k hypervolume AuxSets instead of mid-k/tail-k (ablation) | DoKnowAD · ECG (UCR) | F1-A (%) | Improved |
| increase k beyond a certain value (ablation) | DoKnowAD · ECG (UCR) | F1-A (%) | Hurt |
| plug DoKnowAD into other representation backbones (ablation) | DoKnowAD · other backbones (ECG) | F1-A (%) | Improved |
Closest papers by meaning
- Structured Temporal Causality for Interpretable Multivariate Time Series Anomaly Detection (NeurIPS 2025 · similarity 0.91)
- The Elephant in the Room: Towards A Reliable Time-Series Anomaly Detection Benchmark (NeurIPS 2024 · similarity 0.90)
- Supervision Recovery for Time Series Anomaly Detection via Counterfactual Pairing (NeurIPS 2026 · similarity 0.90)
- Detect, Explain, Interpret: An End-to-End Benchmark for Time Series Anomaly Detection, Explainability and Interpretability. (NeurIPS 2026 · similarity 0.89)
- Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection (NeurIPS 2026 · similarity 0.88)
- CrossAD: Time Series Anomaly Detection with Cross-scale Associations and Cross-window Modeling (NeurIPS 2025 · similarity 0.88)
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.
