ARMove: Learning to Predict Human Mobility through Agentic Reasoning

TL;DR

ARMove introduces an agentic reasoning framework for human mobility prediction, combining interpretability, iterative learning, and transferability, outperforming state-of-the-art methods across diverse datasets.

Contribution

It presents a novel, fully transferable mobility prediction framework leveraging agentic reasoning, feature management, and large-small model synergy for improved accuracy and interpretability.

Findings

Outperforms baselines on 6 out of 12 metrics with gains up to 10.47%.

Demonstrates robustness across regions, users, and model scales.

Provides transparent decision-making analysis.

Abstract

Human mobility prediction is a critical task but remains challenging due to its complexity and variability across populations and regions. Recently, large language models (LLMs) have made progress in zero-shot prediction, but existing methods suffer from limited interpretability (due to black-box reasoning), lack of iterative learning from new data, and poor transferability. In this paper, we introduce \textbf{ARMove}, a fully transferable framework for predicting human mobility through agentic reasoning. To address these limitations, ARMove employs standardized feature management with iterative optimization and user-specific customization: four major feature pools for foundational knowledge, user profiles for segmentation, and an automated generation mechanism integrating LLM knowledge. Robust generalization is achieved via agentic decision-making that adjusts feature weights to maximize accuracy while providing interpretable decision paths. Finally, large-small model synergy distills strategies from large LLMs (e.g., 72B) to smaller ones (e.g., 7B), reducing costs and enhancing performance ceilings. Extensive experiments on four global datasets show ARMove outperforms state-of-the-art baselines on 6 out of 12 metrics (gains of 0.78\% to 10.47\%), with transferability tests confirming robustness across regions, users, and scales. The other 4 items also achieved suboptimal results. Transferability tests confirm its 19 robustness across regions, user groups, and model scales, while interpretability 20 analysis highlights its transparency in decision-making. Our codes are available at: https://anonymous.4open.science/r/ARMove-F847.