Research map / Time series / Anomaly detection
TeamWork: Multivariate Time Series Anomaly Detection via Asymmetric Role-aware Channel Modeling
Shiyan Hu, Tengxue Zhang, Jianxin Jin, Xiangfei Qiu, Bin Yang, Chenjuan Guo
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 dominant/auxiliary role-aware channel modeling instead of channel-independent modeling (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.864 -> 0.927 | Improved |
| add role-aware channel modeling instead of channel-dependent modeling (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.899 -> 0.927 | Improved |
| include dominant variables with their own branch (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.673 -> 0.927 | Improved |
| include auxiliary variables with a lightweight branch (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.902 -> 0.927 | Improved |
| assign roles by entropy reduction index instead of random assignment (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.867 -> 0.927 | Improved |
| use gated interaction instead of simple add fusion (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.889 -> 0.927 | Improved |
| add dominant-guided extraction block (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.908 -> 0.927 | Improved |
| add interaction gate layer (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.904 -> 0.927 | Improved |
| let auxiliary variables also guide dominant representation (bidirectional) (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.927 -> 0.917 | No clear effect |
| model auxiliary variables with the same period-aware masked modeling as dominant ones (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.927 -> 0.905 | No clear effect |
| use period-aware maskers instead of a fixed single masker (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.918 -> 0.927 (vs segment masker only) | Improved |
| add maskers to the self-supervised reconstruction task (ablation) | TeamWork · Credit, Genesis, SMD (avg) | Affiliated-F1: 0.890 -> 0.927 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CATCH — CATCH: Channel-Aware Multivariate Time Series Anomaly Detection via Frequency Patching (ICLR 2025) · better on 11 of 12 shared results
- CrossAD — CrossAD: Time Series Anomaly Detection with Cross-scale Associations and Cross-window Modeling (NeurIPS 2025) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Structured Temporal Causality for Interpretable Multivariate Time Series Anomaly Detection (NeurIPS 2025 · similarity 0.88)
- CATCH: Channel-Aware Multivariate Time Series Anomaly Detection via Frequency Patching (ICLR 2025 · similarity 0.88)
- SARAD: Spatial Association-Aware Anomaly Detection and Diagnosis for Multivariate Time Series (NeurIPS 2024 · similarity 0.87)
- The Elephant in the Room: Towards A Reliable Time-Series Anomaly Detection Benchmark (NeurIPS 2024 · similarity 0.87)
- Channel Matters: Estimating Channel Influence for Multivariate Time Series (NeurIPS 2025 · similarity 0.87)
- Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection (NeurIPS 2026 · similarity 0.86)
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.
