ActivityEditor: Learning to Synthesize Physically Valid Human Mobility

TL;DR

ActivityEditor introduces a dual-LLM-agent framework for zero-shot, physically valid human mobility trajectory synthesis across regions, addressing data scarcity in urban modeling.

Contribution

It presents a novel two-stage framework combining intention generation and trajectory refinement, trained via reinforcement learning for high-fidelity, generalizable mobility simulation.

Findings

Achieves superior zero-shot transfer across diverse urban contexts.

Maintains high statistical fidelity and physical validity in generated trajectories.

Outperforms existing methods in mobility simulation with limited data.

Abstract

Human mobility modeling is indispensable for diverse urban applications. However, existing data-driven methods often suffer from data scarcity, limiting their applicability in regions where historical trajectories are unavailable or restricted. To bridge this gap, we propose \textbf{ActivityEditor}, a novel dual-LLM-agent framework designed for zero-shot cross-regional trajectory generation. Our framework decomposes the complex synthesis task into two collaborative stages. Specifically, an intention-based agent, which leverages demographic-driven priors to generate structured human intentions and coarse activity chains to ensure high-level socio-semantic coherence. These outputs are then refined by editor agent to obtain mobility trajectories through iteratively revisions that enforces human mobility law. This capability is acquired through reinforcement learning with multiple rewards grounded in real-world physical constraints, allowing the agent to internalize mobility regularities and ensure high-fidelity trajectory generation. Extensive experiments demonstrate that \textbf{ActivityEditor} achieves superior zero-shot performance when transferred across diverse urban contexts. It maintains high statistical fidelity and physical validity, providing a robust and highly generalizable solution for mobility simulation in data-scarce scenarios. Our code is available at: https://anonymous.4open.science/r/ActivityEditor-066B.