EUVIMEDEUVIMED entdecken
Crossref

Bridging Cross-Domain Time Series: Efficient Federated Anomaly Detection With Sharded LLMs

Ronghui Xu, Hao Miao, Senzhang Wang, Yan Zhao, Bin Yang, Yunjun Gao, Philip S. Yu, Christian S. Jensen

IEEE Transactions on Knowledge and Data Engineering · 2026-06 · 38 (6) · 3416-3429

DOI: 10.1109/tkde.2026.3677275 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Knowledge and Data Engineering, 2026-06. Verfasst von Ronghui Xu und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 57 Literaturangaben. Gefördert von: National Natural Science Foundation of China, DFF Inge Lehmann, NSF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Time series anomaly detection aims to identify samples that deviate from a normal distribution in a time series, which is practically important to a variety of real-world applications. Existing approaches are mostly centralized and domain-specific, and thus they are hard to generalize to time series of different domains that are decentralized due to the privacy concerns and the resulting data silos across institutions. To bridge this gap, we propose FAST-MAD, the first resource-aware framework for efficient federated time series anomaly detection. Operating under a client-server architecture, different clients in FAST-MAD can handle time series from distinct domains. In particular, FAST-MAD first employs a multi-resolution transformation module to capture hierarchical local semantics, frequency-oriented patching as well as inter-time-series interaction. An LLM serves as the main body of the local model for each client, owing to its strong knowledge transfer capabilities. Further, an adaptive modularized separation mechanism is integrated with sharded federated training to reduce computational costs, which innovatively splits the LLM into a U-shaped architecture. To address data heterogeneity across different clients, we propose a decomposed client-server alignment mechanism, featuring a tailored low-rank parameter decomposition that extracts domain-common knowledge. Extensive experiments on multiple cross-domain time series datasets offer insight into the effectiveness and efficiency of FAST-MAD, which outperforms SOTA baselines by up to 10.25% in terms of F1-score and reduces the training time by 40.93%.

Förderung

  • National Natural Science Foundation of China
  • DFF Inge Lehmann
  • NSF

Zitieren

APA

Xu, R., Miao, H., Wang, S., Zhao, Y., Yang, B., Gao, Y., Yu, P. S., & Jensen, C. S. (2026). Bridging Cross-Domain Time Series: Efficient Federated Anomaly Detection With Sharded LLMs. IEEE Transactions on Knowledge and Data Engineering, 38(6), 3416-3429. https://doi.org/10.1109/tkde.2026.3677275

Vancouver

Xu R, Miao H, Wang S, Zhao Y, Yang B, Gao Y, et al. Bridging Cross-Domain Time Series: Efficient Federated Anomaly Detection With Sharded LLMs. IEEE Transactions on Knowledge and Data Engineering. 2026;38(6):3416-3429. doi:10.1109/tkde.2026.3677275

BibTeX
@article{Xu2026,
  author = {Xu, Ronghui and Miao, Hao and Wang, Senzhang and Zhao, Yan and Yang, Bin and Gao, Yunjun and Yu, Philip S. and Jensen, Christian S.},
  title = {Bridging Cross-Domain Time Series: Efficient Federated Anomaly Detection With Sharded LLMs},
  journal = {IEEE Transactions on Knowledge and Data Engineering},
  year = {2026},
  volume = {38},
  number = {6},
  pages = {3416--3429},
  doi = {10.1109/tkde.2026.3677275}
}
Kostenlos starten →