Refining Time Series Anomaly Detectors using Large Language Models
This work addresses the need for human oversight in anomaly detection systems across industries like finance and healthcare, but it is incremental as it builds on existing LLM capabilities.
The paper tackled the problem of reducing human effort in time series anomaly detection by using multimodal large language models to identify false alarms, achieving effective results through integration of visual and textual data.
Time series anomaly detection (TSAD) is of widespread interest across many industries, including finance, healthcare, and manufacturing. Despite the development of numerous automatic methods for detecting anomalies, human oversight remains necessary to review and act upon detected anomalies, as well as verify their accuracy. We study the use of multimodal large language models (LLMs) to partially automate this process. We find that LLMs can effectively identify false alarms by integrating visual inspection of time series plots with text descriptions of the data-generating process. By leveraging the capabilities of LLMs, we aim to reduce the reliance on human effort required to maintain a TSAD system