The paper introduces COLA (Cross-city Mobility Transformer), a novel method for simulating human trajectories across cities. The authors address the challenge of data scarcity and domain heterogeneity in human trajectory simulation by leveraging external mobility data. COLA is designed to transfer universal patterns of human mobility from abundant external cities to improve the synthetic quality of trajectories in the target city. The method divides the Transformer into private modules for city-specific characteristics and shared modules for city-universal mobility patterns. It also employs a post-hoc adjustment strategy to calibrate the prediction probabilities of locations, addressing the overconfident problem in deep models. Extensive experiments on four real-world city datasets demonstrate the superiority of COLA over state-of-the-art single-city and cross-city baselines. The code for COLA is available at https://github.com/Star607/Cross-city-Mobility-Transformer.The paper introduces COLA (Cross-city Mobility Transformer), a novel method for simulating human trajectories across cities. The authors address the challenge of data scarcity and domain heterogeneity in human trajectory simulation by leveraging external mobility data. COLA is designed to transfer universal patterns of human mobility from abundant external cities to improve the synthetic quality of trajectories in the target city. The method divides the Transformer into private modules for city-specific characteristics and shared modules for city-universal mobility patterns. It also employs a post-hoc adjustment strategy to calibrate the prediction probabilities of locations, addressing the overconfident problem in deep models. Extensive experiments on four real-world city datasets demonstrate the superiority of COLA over state-of-the-art single-city and cross-city baselines. The code for COLA is available at https://github.com/Star607/Cross-city-Mobility-Transformer.