TimeThink: Eliciting Compositional Reasoning in Timeseries Large Language Models

2026-09-11

AI Summary

为解决时间序列模型在动态模式捕捉和解释性上的不足，TimeThink通过合成数据生成器与强化学习结合的方法提升模型的组合推理能力。

Abstract

Timeseries multimodal large language models (TS-MLLMs) have recently begun leveraging the reasoning capabilities of large language models (LLMs) for question-answering tasks. However, these models often fail to capture dynamic temporal patterns, providing only implicit reasoning that lacks the underlying explanations critical for high-stakes applications like healthcare. While reinforcement learning (RL)-based timeseries language models aim to address this, they often fall short because they are trained on narrow, in-distribution data and struggle with out-of-distribution compositional questions. To address these challenges, we present TimeThink, a synthetic framework for eliciting compositional timeseries reasoning. Core timeseries primitives (e.g., trend, seasonality) are domain-independent and can be deterministically generated. Guided by this premise, TimeThink first designs a synthetic data generator that produces atomic and composite question-answer pairs, providing objective ground truth with reasoning traces. Building on this framework, TimeThink employs a reinforcement learning with verifiable rewards (RLVR) training strategy that encourages explicit reasoning. Unlike template-reliant methods, this approach enables the model to learn the underlying logic of composition rather than simply imitating traces. Extensive experiments show that TimeThink, trained only on synthetic data, significantly outperforms strong baselines on both synthetic and real-world benchmarks.

Problem

• Timeseries Multimodal Large Language Models
• Dynamic Temporal Patterns
• Reinforcement Learning
• Out-of-Distribution Compositional Questions

Innovation

• synthetic framework
• compositional reasoning
• reinforcement learning with verifiable rewards (RLVR)
• synthetic data generator