AI RESEARCH

When Normality Shifts: Risk-Aware Test-Time Adaptation for Unsupervised Tabular Anomaly Detection

arXiv CS.AI

ArXi:2605.10242v1 Announce Type: cross Unsupervised tabular anomaly detection methods typically and subsequently identify samples that deviate from these patterns as anomalies during testing. However, in practical scenarios, the limited scale and diversity of