返回全部动态
将聚合移动统计蒸馏为语言模型策略用于赛后人群模拟
原标题:Distilling Aggregate Mobility Statistics into a Language Model Policy for Post-Event Crowd Simulation
AI 摘要
该研究提出一种方法,通过微调语言模型智能体,使人群模拟中的目的地分布与从聚合移动数据中观测到的OD流一致。研究者使用迭代比例拟合和重采样校正训练数据,避免了主导类别的过度膨胀。在棒球比赛数据上的实验表明,该方法将目的地份额误差降低了25%,且无需推理时校正。
以上摘要由 AI 生成,可能存在误差。事实请以原文为准。
正文节选
Distilling Aggregate Mobility Statistics into a Language Model Policy for Post-Event Crowd SimulationConference: the 34th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems; November 3–6, 2026; Riverside, CA, USACCS: Information systems Geographic information systemsCCS: Computing methodologies Multi-agent systemsCCS: Computing methodologies Artificial intelligence Abstract. Pedestrian simulators need a behaviour rule for every agent, but privacy usually limits
发布时间:2026-08-21 12:00
抓取时间:2026-08-21 12:03
来源机构:arXiv