CastFlow: Learning Role-Specialized Agentic Workflows for Time Series Forecasting

ArXi:2604.27840v1 Announce Type: cross Recently, large language models (LLMs) have shown great promise in time series forecasting. However, most existing LLM-based forecasting methods still follow a static generative paradigm that directly maps historical observations to future values in a single pass. Under this paradigm, forecasting is constrained by limited temporal pattern extraction, single-round acquisition of contextual features, one-shot forecast generation, and lack of from ensemble forecasts.