Harnessing Vision-Language Models for Time Series Anomaly Detection
Zelin He, Sarah Alnegheimish, Matthew Reimherr
Abstract
Time-series anomaly detection (TSAD) has played a vital role in a variety of fields, including healthcare, finance, and sensor-based condition monitoring. Prior methods, which mainly focus on training domain-specific models on numerical data, lack the visual–temporal reasoning capacity that human experts have to identify contextual anomalies. To fill this gap, we explore a solution based on vision language models (VLMs). Recent studies have shown the ability of VLMs for visual reasoning tasks, yet their direct application to time series has fallen short on both accuracy and efficiency. To harness the power of VLMs for TSAD, we propose a two-stage solution, with (1) ViT4TS, a vision-screening stage built on a relatively lightweight pre-trained vision encoder, which leverages 2-D time series representations to accurately localize candidate anomalies; (2) VLM4TS, a VLM-based stage that integrates global temporal context and VLM
BibTeX
@inproceedings{aaai2026_harnessingvision,
title = {Harnessing Vision-Language Models for Time Series Anomaly Detection},
author = {Zelin He and Sarah Alnegheimish and Matthew Reimherr},
booktitle = {AAAI 2026},
year = {2026}
}