CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (17) · 15018-15026
DOI: 10.1609/aaai.v40i17.38524 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von HyunGi Kim und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref).
Abstract (Originalsprache)
Multivariate time-series anomaly detection (MTSAD) aims to identify deviations from normality in multivariate time-series and is critical in real-world applications. However, in real-world deployments, distribution shifts are ubiquitous and cause severe performance degradation in pre-trained anomaly detector. Test-time adaptation (TTA) updates a pre-trained model on-the-fly using only unlabeled test data, making it promising for addressing this challenge. In this study, we propose CANDI (Curated test-time adaptation for multivariate time-series ANomaly detection under DIstribution shift), a novel TTA framework that selectively identifies and adapts to potential false positives while preserving pre-trained knowledge. CANDI introduces a False Positive Mining (FPM) strategy to curate adaptation samples based on anomaly scores and latent similarity, and incorporates a plug-and-play Spatiotemporally-Aware Normality Adaptation (SANA) module for structurally informed model updates. Extensive experiments demonstrate that CANDI significantly improves the performance of MTSAD under distribution shift, improving AUROC up to 14% while using fewer adaptation samples.
Zitieren
APA
Kim, H., Mok, J., Lee, H., Shin, J., & Yoon, S. (2026). CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift. Proceedings of the AAAI Conference on Artificial Intelligence, 40(17), 15018-15026. https://doi.org/10.1609/aaai.v40i17.38524
Vancouver
Kim H, Mok J, Lee H, Shin J, Yoon S. CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(17):15018-15026. doi:10.1609/aaai.v40i17.38524
BibTeX
@article{Kim2026,
author = {Kim, HyunGi and Mok, Jisoo and Lee, Hyungyu and Shin, Juhyeon and Yoon, Sungroh},
title = {CANDI: Curated Test-Time Adaptation for Multivariate Time-Series Anomaly Detection Under Distribution Shift},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {17},
pages = {15018--15026},
doi = {10.1609/aaai.v40i17.38524}
}
EUVIMED