PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection

ArXi:2602.01359v2 Announce Type: replace Although recent studies on time-series anomaly detection have increasingly adopted ever-larger neural network architectures such as transformers and foundation models, they incur high computational costs and memory usage, making them impractical for real-time and resource-constrained scenarios. Moreover, they often fail to nstrate significant performance gains over simpler methods under rigorous evaluation protocols.