让时间序列预测像智能体一样思考、反馈、迭代，突破传统模型局限。
Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting
- 将预测重构为感知-规划-行动-反思-记忆的智能体流程
- 支持工具调用、结果反馈与持续经验积累，实现动态适应
- 适合需要长期推理与交互的复杂场景，如金融、能源调度
时间序列预测传统上被视为以模型为中心、静态且单次完成的预测任务，将历史观测映射到未来值。尽管这一范式推动了显著进展，但在需要自适应与多轮交互的场景中表现不足，难以实现信息提取、推理驱动推断、迭代优化和持续适应。本文提出代理式时间序列预测（ATSF），将预测重新定义为包含感知、规划、行动、反思与记忆的代理过程。不同于仅关注预测模型，ATSF强调构建可与工具交互、吸收结果反馈并随经验演进的预测工作流。我们提出三种典型实现范式：基于工作流的设计、代理强化学习，以及混合代理工作流范式，并讨论从模型中心预测转向代理预测带来的机遇与挑战。本文旨在确立代理预测作为时序预测未来研究的基础。
原文摘要 · Abstract (English)
Time series forecasting has traditionally been formulated as a model-centric, static, and single-pass prediction problem that maps historical observations to future values. While this paradigm has driven substantial progress, it proves insufficient in adaptive and multi-turn settings where forecasting requires informative feature extraction, reasoning-driven inference, iterative refinement, and continual adaptation over time. In this paper, we argue for agentic time series forecasting (ATSF), which reframes forecasting as an agentic process composed of perception, planning, action, reflection, and memory. Rather than focusing solely on predictive models, ATSF emphasizes organizing forecasting as an agentic workflow that can interact with tools, incorporate feedback from outcomes, and evolve through experience accumulation. We outline three representative implementation paradigms -- workflow-based design, agentic reinforcement learning, and a hybrid agentic workflow paradigm -- and discuss the opportunities and challenges that arise when shifting from model-centric prediction to agentic forecasting. Together, this position aims to establish agentic forecasting as a foundation for future research at the intersection of time series forecasting.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。