A Transformer with global dictionary cross-attention and prototype similarity achieves state-of-the-art unsupervised multivariate time series anomaly detection on standard benchmarks.
Specifically, compared with DCdetector, GDformer can promise 57.32% F1-score improvements on NIPS_TS_GECCO dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GDformer: Going Beyond Subsequence Isolation for Multivariate Time Series Anomaly Detection
A Transformer with global dictionary cross-attention and prototype similarity achieves state-of-the-art unsupervised multivariate time series anomaly detection on standard benchmarks.