LLMs generate human mobility patterns accounting for encounters

Researchers have developed EncounterMob, a novel framework utilizing large language models (LLMs) to generate human mobility patterns that account for real-world encounters between individuals. Unlike previous resident-centric approaches, EncounterMob jointly models individual trajectories and population-level interaction structures. The framework employs a coarse-to-fine Encounter Generator and an Iterative Coupling Framework to coordinate encounters with personal mobility behaviors. Experiments on datasets from Japan and New York City demonstrate that EncounterMob significantly outperforms existing methods in both encounter and personal mobility metrics. AI

IMPACT This research could lead to more realistic simulations for urban planning and public health by better modeling social interactions.

RANK_REASON Academic paper detailing a new method for LLM-based human mobility generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• arXiv
• ELLMob
• EncounterMob
• Hugging Face
• Japan
• large-language models
• New York City

• paper
• model release

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