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

arXiv: https://arxiv.org/abs/2604.27840 · HTML: https://mlsanity.com/paper/2604.27840/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Financial trading and forecasting agents, one of 16 directions in LLM agents and tool use.
- Builds on 14 of its 71 references that are in the map.
- First posted on arXiv 30 Apr 2026.

## Placed in

- LLM agents and tool use / [Financial trading and forecasting agents](https://mlsanity.com/agents/financial-trading-and-forecasting-agents/)

## Builds on (most cited first, 14 of 14)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Deep Time Series Models: A Comprehensive Survey and Benchmark](https://mlsanity.com/paper/2407.13278/) (2024)
- [Agent Lightning: Train ANY AI Agents with Reinforcement Learning](https://mlsanity.com/paper/2508.03680/) (2025)
- [Time Series Forecasting via Reasoning: A Slow-Thinking Approach with Reinforcement Fine-Tuned LLMs](https://mlsanity.com/paper/2506.10630/) (2025)
- [TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis](https://mlsanity.com/paper/2510.01538/) (2025)
- [Improving Time Series Forecasting via Instance-aware Post-hoc Revision](https://mlsanity.com/paper/2505.23583/) (2025)
- [From Values to Tokens: An LLM-Driven Framework for Context-aware Time Series Forecasting via Symbolic Discretization](https://mlsanity.com/paper/2508.09191/) (2025)
- [AlphaCast: A Human Wisdom-LLM Intelligence Co-Reasoning Framework for Interactive Time Series Forecasting](https://mlsanity.com/paper/2511.08947/) (2025)
- [Cast-R1: Learning Tool-Augmented Sequential Decision Policies for Time Series Forecasting](https://mlsanity.com/paper/2602.13802/) (2026)
- [Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting](https://mlsanity.com/paper/2602.01776/) (2026)
- [AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoning](https://mlsanity.com/paper/2602.13807/) (2026)
- [One-Embedding-Fits-All: Efficient Zero-Shot Time Series Forecasting by a Model Zoo](https://mlsanity.com/paper/2509.04208/) (2025)
- [InstructTime++: Time Series Classification with Multimodal Language Modeling via Implicit Feature Enhancement](https://mlsanity.com/paper/2601.14968/) (2026)

## Built on by (most cited first, 2 of 2)

- [Memory in Deep Time-Series Models](https://mlsanity.com/paper/2609.06006/) (2026)
- [When Tomorrow Becomes Today: Self-Evolving Policies for Agentic Time-Series Forecasting](https://mlsanity.com/paper/2609.24862/) (2026)

## Abstract (arXiv)

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 support from ensemble forecasts. To address these limitations, in this work, we propose CastFlow, a dynamic agentic forecasting framework that enables multi-view temporal pattern extraction, multi-round contextual features acquisition, iterative forecast refinement, and forecasting with ensemble forecasts. First, CastFlow organizes the forecasting process into planning, action, forecasting, and reflection, establishing an agentic workflow. Second, this workflow is supported by a memory module that retrieves prior experience and a multi-view toolkit that constructs diagnostic evidence and provides a reliable ensemble forecast baseline. Third, CastFlow adopts a role-specialized design that combines general-purpose reasoning with specialized numerical forecasting. Under this design, a frozen LLM preserves general-purpose reasoning, while a fine-tuned domain-specific LLM performs evidence-guided numerical forecasting based on the ensemble forecast baseline, rather than from scratch. To optimize a fine-tuned domain-specific LLM, we further develop a two-stage workflow-oriented training that combines supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR). To evaluate the effectiveness of CastFlow, we conduct extensive experiments on diverse datasets and show that it achieves superior overall results against strong baselines. We hope that this work can serve as a step toward more adaptive and accurate time series forecasting.