2026 · Survey preprint
The Landscape of Agentic Time Series Systems: Architectures, Reliability, and Frontiers
Yifan Hu, Jie Yang, Xilin Dai, Wanxu Cai, Kuiye Ding, Yuante Li, Qinghua Liu, Enze Ma, Zhiyuan Qu, Yixin Wang, Binyan Xu, Kexin Zhang, Peiyuan Liu, Zhijian Xu, Guibin Zhang, Yujin Tang, Yanwei Yue, Kening Zheng, Chengze Li, Hanrong Zhang, Haoyan Xu, Naiqi Li, Tao Dai, Dawei Cheng, John Paparrizos, Kaize Ding, Tian Zhou, Qiang Xu, Shu-tao Xia, Shirui Pan, Philip S. Yu
Overview
This survey examines how time-series analysis is developing into systems that gather evidence, choose tools or actions, receive feedback, and adapt. It organizes the literature around perception, reasoning, planning and action, memory and knowledge, and temporal world models. Reliability and evaluation connect these capabilities, with particular attention to grounding, reproducibility, and decision-making under uncertainty.
Resources