UniMove: A Unified Model for Multi-city Human Mobility Prediction

2025-08-09

AI Summary

Modeling heterogeneity across cities—arising from disparities in urban structure, infrastructure, and population density—hinders generalization in multi-city human mobility prediction. Method: We propose the first unified cross-city prediction framework, featuring a trajectory-location dual-tower architecture to learn city-agnostic spatial representations; a Mixture-of-Experts (MoE) Transformer to adaptively capture diverse mobility patterns; and cross-city sequence modeling with data augmentation to enable joint multi-city training. Contribution/Results: Evaluated on multiple real-world city datasets, our model achieves an average 10.2% improvement in prediction accuracy over single-city baselines. It demonstrates significantly enhanced generalization and cross-city transferability, establishing a novel paradigm for universal spatiotemporal modeling in urban computing.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Human mobility prediction is vital for urban planning, transportation optimization, and personalized services. However, the inherent randomness, non-uniform time intervals, and complex patterns of human mobility, compounded by the heterogeneity introduced by varying city structures, infrastructure, and population densities, present significant challenges in modeling. Existing solutions often require training separate models for each city due to distinct spatial representations and geographic coverage. In this paper, we propose UniMove, a unified model for multi-city human mobility prediction, addressing two challenges: (1) constructing universal spatial representations for effective token sharing across cities, and (2) modeling heterogeneous mobility patterns from varying city characteristics. We propose a trajectory-location dual-tower architecture, with a location tower for universal spatial encoding and a trajectory tower for sequential mobility modeling. We also design MoE Transformer blocks to adaptively select experts to handle diverse movement patterns. Extensive experiments across multiple datasets from diverse cities demonstrate that UniMove truly embodies the essence of a unified model. By enabling joint training on multi-city data with mutual data enhancement, it significantly improves mobility prediction accuracy by over 10.2%. UniMove represents a key advancement toward realizing a true foundational model with a unified architecture for human mobility. We release the implementation at https://github.com/tsinghua-fib-lab/UniMove/.

Problem

• Develop unified model for multi-city mobility prediction
• Address heterogeneous city structures and patterns
• Improve prediction accuracy with joint training

Innovation

• Unified model for multi-city mobility prediction
• Trajectory-location dual-tower architecture design
• MoE Transformer blocks for diverse patterns