Rationale-Grounded In-Context Learning for Time Series Reasoning with Multimodal Large Language Models

2026-01-06

arXiv.org

AI Summary

Existing multimodal large language models often rely on superficial pattern matching for time series reasoning, lacking the principled ability to connect temporal observations with downstream outcomes. To address this limitation, this work proposes RationaleTS, a rationale-based in-context learning framework that uniquely treats reasoning pathways as proactive guidance rather than post-hoc explanations. RationaleTS generates structured rationales conditioned on labels and introduces a hybrid retrieval mechanism that integrates temporal patterns with semantic context to incorporate relevant reasoning priors. Evaluated across time series tasks in three distinct domains, RationaleTS significantly outperforms current baselines, demonstrating its effectiveness and efficiency in enhancing the model’s capacity for principled, causal-style reasoning.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Time-Series/Data Streams
• Reasoning under Uncertainty: Causality

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The underperformance of existing multimodal large language models for time series reasoning lies in the absence of rationale priors that connect temporal observations to their downstream outcomes, which leads models to rely on superficial pattern matching rather than principled reasoning. We therefore propose the rationale-grounded in-context learning for time series reasoning, where rationales work as guiding reasoning units rather than post-hoc explanations, and develop the RationaleTS method. Specifically, we firstly induce label-conditioned rationales, composed of reasoning paths from observable evidence to the potential outcomes. Then, we design the hybrid retrieval by balancing temporal patterns and semantic contexts to retrieve correlated rationale priors for the final in-context inference on new samples. We conduct extensive experiments to demonstrate the effectiveness and efficiency of our proposed RationaleTS on three-domain time series reasoning tasks. We will release our code for reproduction.

Problem

• time series reasoning
• multimodal large language models
• rationale priors
• in-context learning

Innovation

• rationale-grounded reasoning
• in-context learning
• time series reasoning
• multimodal large language models
• hybrid retrieval