Evaluating Large Language Models on Time Series Feature Understanding: A Comprehensive Taxonomy and Benchmark

arXiv:2404.16563v1 Announce Type: new
Abstract: Large Language Models (LLMs) offer the potential for automatic time series analysis and reporting, which is a critical task across many domains, spanning healthcare, finance, climate, energy, and many more. In this paper, we propose a framework for rigorously evaluating the capabilities of LLMs on time series understanding, encompassing both univariate and multivariate forms. We introduce a comprehensive taxonomy of time series features, a critical framework that delineates various characteristics inherent in time series data. …

abstract analysis arxiv benchmark climate cs.cl domains energy feature finance framework healthcare language language models large language large language models llms paper reporting series taxonomy time series type understanding