Bridging Classification and Reconstruction: Cooperative Time Series Anomaly Detection

In three linesCoAD, a novel framework for time series anomaly detection, unifies classification (Outlier Exposure) and reconstruction (Masked Autoencoder) paradigms. The classification module generates probability-informed soft masks for the reconstruction module, addressing generalization and masking misalignment issues. Experiments on standard benchmarks demonstrate significant improvements with faster inference.

Summary generated by Claude — human-verified