Language models fine-tuned for crowd simulation using aggregate mobility data

Researchers have developed a method to fine-tune language models for crowd simulation by using aggregate mobility statistics. This approach addresses the challenge of under-determined individual behavior when only aggregate data like zone-level device counts and origin-to-destination flows are available. By iteratively fitting the model's destination distribution to observed data, the fine-tuned agent achieved a 25% reduction in destination-share error for two baseball games, while maintaining similar grid correlation. AI

IMPACT This research could improve the accuracy and efficiency of crowd simulation models, impacting urban planning and event management.

RANK_REASON The cluster contains an academic paper detailing a new method for using language models in crowd simulation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• crowd simulation
• Distilling Aggregate Mobility Statistics into a Language Model Policy for Post-Event Crowd Simulation
• Hugging Face
• language model

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →