Abstract
This paper presents an enhanced Anomaly Transformer for unsupervised anomaly detection in multivariate time series. We introduce a contrastive embedding module that refines the input embedding by optimizing a lightweight MLP with an InfoNCE objective, where positive pairs are pulled closer while negative pairs are pushed apart in the feature space. We further propose Unified Discrimination at the decision stage, which unifies reconstruction error and association discrepancy into a single anomaly score. Experiments on four large-scale benchmarks (SMD, PSM, MSL, and SMAP) demonstrate consistent performance gains over the Anomaly Transformer baseline, improving Precision and F1-score while maintaining high Recall, with an average F1-score of 0.9554.
| Original language | English |
|---|---|
| Title of host publication | 2026 IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) |
| Publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| Pages | 821-822 |
| Number of pages | 2 |
| ISBN (Electronic) | 9798319506429 |
| ISBN (Print) | 9798319506436 (PoD) |
| DOIs | |
| Publication status | Published - 24 Aug 2026 |
| Event | 13th IEEE International Conference on Consumer Electronics - Taiwan - Taoyuan, Taiwan, Province of China Duration: 1 Jul 2026 → 3 Jul 2026 |
Publication series
| Name | IEEE International Conference on Consumer Electronics-Taiwan |
|---|---|
| Publisher | IEEE |
| ISSN (Print) | 2381-5779 |
Conference
| Conference | 13th IEEE International Conference on Consumer Electronics - Taiwan |
|---|---|
| Abbreviated title | ICCE-Taiwan 2026 |
| Country/Territory | Taiwan, Province of China |
| City | Taoyuan |
| Period | 1/07/26 → 3/07/26 |
Bibliographical note
Publisher Copyright:© 2026 IEEE.
Keywords
- anomaly detection
- contrastive embedding
- Time-series data
- Transformer
- Unified Discrimination
ASJC Scopus subject areas
- Human-Computer Interaction
- Electrical and Electronic Engineering
- Media Technology
- Modelling and Simulation
- Instrumentation
Fingerprint
Dive into the research topics of 'Enhanced Transformer Powered Time-Series Anomaly Detection via Contrastive Embedding and Unified Discrimination'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver