用LLM辅助模拟城市行为，10万市民75天仿真一小时搞定
CityBehavEx: A Scalable and Empirically Validated LLM-Assisted Urban Simulation Platform
- 融合经典移动模型与微调交叉编码器，降低大規模模擬成本
- 10万代理人在单张消费级显卡上75天仿真耗时不足1小时
- 支持行为可追溯、结果可验证，适配城市规划与交通研究
基于大语言模型的多智能体城市模拟系统虽能生成语义丰富的城市活动，但存在扩展性差、实证验证弱的问题。我们提出CityBehavEx——一个可交互、可扩展的LLM辅助城市模拟平台，支持百万级规模的城市人口仿真，能暴露代理行为以供检查，支持对真实世界出行模式的实证验证，并生成更符合真实时空与语义分布的移动模式。该平台不为每个代理动作调用大模型，而是结合成熟的人类移动模型与微调的交叉编码器，用于估算代理特征、日程安排与活动转换之间的语义一致性。在一项案例研究中，10万代理在75天内的仿真仅用不到一小时即可完成（单个消费级GPU）。用户可定义模拟区域、启动实验、查看轨迹与活动记录、调试不合理行为，并基于真实世界的出行、时间使用和语义指标验证生成结果。
原文摘要 · Abstract (English)
Recent LLM-based multi-agent urban simulators can generate semantically rich city routines, but they remain costly to scale and are often weakly validated against empirical mobility patterns. We present CityBehavEx, an interactive LLM-assisted urban simulation platform that scales to city-size populations, exposes agent behavior for inspection, supports empirical validation, and generates mobility patterns that better match real-world spatial, temporal, and semantic distributions. Instead of invoking large language models for every agent action, CityBehavEx combines established human mobility models with fine-tuned cross-encoders that estimate semantic alignment between agent profiles, schedules, and activity transitions. This design enables large-scale simulations, as demonstrated in a case study of 100,000 agents over 75 days in under one hour on a single consumer GPU. The platform allows users to define simulation regions, launch experiments, inspect trajectories and activity traces, debug unrealistic behaviors, and validate generated routines against real-world mobility, time-use, and semantic metrics.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。