TimeThink improves reasoning on temporal patterns for language models
TimeThink: Eliciting Compositional Reasoning in Timeseries Large Language Models
Artificial Intelligence
Summary
Timeseries language models often struggle to explain how they understand changing patterns over time, which is important in areas like healthcare. The authors created TimeThink, a new way to train these models using synthetic data that includes clear reasoning steps. This method helps the models learn to answer complex questions about time-based data more accurately, even when they see new types of questions. Their approach shines both on artificial and real-world tests, showing it can teach better reasoning without relying on narrow datasets.
What this means in practice
- •For healthcare data teams: Improve explainable analysis of patient monitoring timeseries data for critical decision-making.
- •For financial analytics teams: Enhance models that interpret market trends and seasonal patterns with clear reasoning on complex timeseries questions.
Authors
Sudarshan Regmi, Arvind Pillai, Yu Yvonne Wu, Yuliang Chen, Bibek Panthi, Tess Z. Griffin, Michael V. Heinz, Lisa Marsch, Nicholas C. Jacobson, Andrew Campbell
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.