Kadir-Kaan Özer

JuRe reveals that a simple architecture can outperform complex models in time series anomaly detection by focusing on the right training objectives.

Time series anomalies hiding in plain sight? AxonAD spots them by predicting how attention queries *should* evolve, then flags deviations from that expectation.

Accuracy leaderboards mislead: lightweight classical anomaly detectors surprisingly outperform deep methods when deployed under the throughput constraints of in-vehicle monitoring systems.