Mobility-Aware Cache Framework for Scalable LLM-Based Human Mobility Simulation
This work addresses scalability issues for urban planning, epidemiology, and transportation analysis, but it is incremental as it builds on existing LLM-based simulation methods.
The paper tackles the high computational cost limiting scalability in large-scale human mobility simulations using LLMs, and proposes MobCache, a mobility-aware cache framework that significantly improves efficiency while maintaining performance comparable to state-of-the-art methods.
Large-scale human mobility simulation is critical for applications such as urban planning, epidemiology, and transportation analysis. Recent works treat large language models (LLMs) as human agents to simulate realistic mobility behaviors using structured reasoning, but their high computational cost limits scalability. To address this, we design a mobility-aware cache framework named MobCache that leverages reconstructible caches to enable efficient large-scale human mobility simulations. It consists of: (1) a reasoning component that encodes each reasoning step as a latent-space embedding and uses a latent-space evaluator to enable the reuse and recombination of reasoning steps; and (2) a decoding component that employs a lightweight decoder trained with mobility law-constrained distillation to translate latent-space reasoning chains into natural language, thereby improving simulation efficiency while maintaining fidelity. Experiments show that MobCache significantly improves efficiency across multiple dimensions while maintaining performance comparable to state-of-the-art LLM-based methods.