CTRL: Control-Based Time Series Forecasting with LLM-Guided Residual Learning

2026-09-19

AI Summary

针对非平稳时间序列预测不稳定问题，提出CTRL框架，通过冻结骨干生成基础预测，并利用LLM代理分析预测误差，生成校正信号以提高预测鲁棒性。

Abstract

Time series forecasting underpins critical decision-making across diverse domains. While large language models (LLMs) offer promising reasoning capabilities, existing LLM-based time series forecasting approaches either reduce them to numerical predictors that bypass their strengths, or allow direct forecast generation that destabilizes predictions in non-stationary settings. We introduce CTRL, a framework that decouples semantic reasoning from quantitative prediction. A frozen backbone generates base forecasts, while specialized LLM agents function as controllers that analyze backbone prediction errors through decomposed trend, seasonal, and irregular components, grounding reasoning in interpretable temporal structure. Each agent outputs compact control signals that a lightweight residual decoder translates into forecast corrections. CTRL incorporates label-free test-time adaptation that detects distribution shift from input statistics alone and readapts control signals with only 3-24 LLM calls via caching. CTRL is explicitly designed to improve robustness under non-stationary temporal dynamics and distribution shift, while remaining competitive on highly stationary time series where adaptive correction provides limited additional benefit.

Problem

• time series forecasting
• large language models
• non-stationary settings
• distribution shift
• semantic reasoning

Innovation

• Control-Based Time Series Forecasting
• LLM-Guided Residual Learning
• Decoupled Semantic Reasoning
• Test-Time Adaptation
• Non-Stationary Temporal Dynamics