TimeXL: Explainable Multi-modal Time Series Prediction with LLM-in-the-Loop

TL;DR

TimeXL is a multi-modal time series prediction framework that combines prototype-based encoding with large language models to improve accuracy and provide interpretable, human-centric explanations through a closed-loop reasoning process.

Contribution

It introduces a novel multi-modal prediction system integrating LLMs for enhanced accuracy and interpretability in time series analysis, with a unique feedback loop for continuous improvement.

Findings

Achieves up to 8.9% improvement in AUC on real-world datasets.

Provides human-centric, multi-modal explanations.

Demonstrates effective LLM-driven reasoning for time series prediction.

Abstract

Time series analysis provides essential insights for real-world system dynamics and informs downstream decision-making, yet most existing methods often overlook the rich contextual signals present in auxiliary modalities. To bridge this gap, we introduce TimeXL, a multi-modal prediction framework that integrates a prototype-based time series encoder with three collaborating Large Language Models (LLMs) to deliver more accurate predictions and interpretable explanations. First, a multi-modal prototype-based encoder processes both time series and textual inputs to generate preliminary forecasts alongside case-based rationales. These outputs then feed into a prediction LLM, which refines the forecasts by reasoning over the encoder's predictions and explanations. Next, a reflection LLM compares the predicted values against the ground truth, identifying textual inconsistencies or noise. Guided by this feedback, a refinement LLM iteratively enhances text quality and triggers encoder retraining. This closed-loop workflow-prediction, critique (reflect), and refinement-continuously boosts the framework's performance and interpretability. Empirical evaluations on four real-world datasets demonstrate that TimeXL achieves up to 8.9% improvement in AUC and produces human-centric, multi-modal explanations, highlighting the power of LLM-driven reasoning for time series prediction.