Human mobility patterns generated across cities with shared semantic codes
Plan-to-Synthesis: Cross-City Human Mobility Generation via Semantic Latent Flow Matching
Social and Information NetworksMachine Learning
Summary
Predicting how people move between places in different cities helps with travel planning and managing transportation. Prior methods either had to learn each city separately or used costly models that don’t scale well for long trips. The authors created SeMoFlow, which represents places from many cities in a shared code system to capture common patterns. Their method first plans a rough movement pattern and then fills in details to produce realistic trips that match each city's unique traits. Tests show this approach works better across cities and captures meaningful travel behaviors.
What this means in practice
- •For urban planners: Generate realistic multi-city human travel patterns to improve transportation planning and urban infrastructure design.
- •For transportation management teams: Simulate human movement across different cities to optimize transit operations and predict travel demand patterns.
Authors
Zhoufu Wang, Baoshen Guo, Zhiqing Hong, Junyi Li, Kailai Sun, Heye Huang, Alok Prakash, Shenhao Wang, Jinhua Zhao
Abstract
Human mobility generation aims to synthesize realistic point-of-interest (POI) visitation trajectories and has become an important tool for travel behavior modeling, transportation management, and urban planning. Existing diffusion-based methods achieve high fidelity but require per-city generation, given the inherent heterogeneity of geospatial locations and POI categories, while large language model-based methods generalize across cities but remain too costly at scale, especially for long-horizon trajectory generation. To address this, we propose SeMoFlow, a Semantic human Mobility generation framework based on latent Flow matching. We first encode heterogeneous POIs from different cities into a shared cross-city representation space via hierarchical Semantic IDs, where shared prefixes capture transferable semantics, and successive codes progressively refine the representation toward individual POIs. Building on the semantic IDs, SeMoFlow follows a plan-to-synthesis hierarchical generation paradigm, in which an autoregressive planner generates coarse-grained semantic and recurrence patterns, and a flow matching realizer synthesizes fine-grained suffix latents. The generated latents are subsequently decoded and grounded to concrete POIs. Extensive experiments on large-scale multi-city datasets show that SeMoFlow achieves higher trajectory fidelity than existing baselines, preserves city-specific mobility motifs, and supports both joint multi-city generation and effective cross-city transfer.