Adaptive Time Series Reasoning via Segment Selection

2026-02-20

AI Summary

This work addresses the challenge of temporal reasoning over long sequences, where models must locate sparse, task-relevant segments to answer natural language questions—a task poorly handled by conventional approaches that statically encode entire sequences. The authors propose ARTIST, a novel framework that couples adaptive temporal segment selection with reasoning through a controller-reasoner architecture. The controller dynamically selects the most informative segments via reinforcement learning, while the reasoner generates answers based solely on these selected fragments. Both components are jointly trained through hierarchical policy optimization. Evaluated across six benchmarks, ARTIST achieves an average accuracy improvement of 6.46 percentage points over the strongest baseline, demonstrating particularly strong performance in rare event localization and multi-segment reasoning scenarios.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Reasoning under Uncertainty: Sequential Decision Making
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Time series reasoning tasks often start with a natural language question and require targeted analysis of a time series. Evidence may span the full series or appear in a few short intervals, so the model must decide what to inspect. Most existing approaches encode the entire time series into a fixed representation before inference, regardless of whether or not the entire sequence is relevant. We introduce ARTIST, which formulates time-series reasoning as a sequential decision problem. ARTIST interleaves reasoning with adaptive temporal segment selection. It adopts a controller-reasoner architecture and uses reinforcement learning to train the controller role to select informative segments and the reasoner role to generate segment-conditioned reasoning traces and final answers. During inference, the model actively acquires task-relevant information instead of relying on a static summary of the full sequence. We use a novel hierarchical policy optimization approach for post-training that allows the model to excel in both segment selection and question-answering behavior. We evaluate ARTIST on six time-series reasoning benchmarks and compare it with large language models, vision-language models, and prior time-series reasoning systems. ARTIST improves average accuracy by 6.46 absolute percentage points over the strongest baseline. The largest gains appear on rare event localization and multi-segment reasoning tasks. Supervised fine-tuning improves performance, and reinforcement learning provides additional gains by optimizing question-adaptive segment selection. These results show that selective data use drives effective time-series reasoning.

Problem

• time series reasoning
• segment selection
• adaptive reasoning
• rare event localization
• multi-segment reasoning

Innovation

• adaptive segment selection
• time series reasoning
• reinforcement learning
• controller-reasoner architecture
• hierarchical policy optimization