CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift

CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift explores A novel test-time adaptation framework for multivariate time-series anomaly detection that significantly improves performance under distribution shifts by selectively adapting to false positives.. Commercial viability score: 7/10 in Time-Series Anomaly Detection.