CityBehavEx is an LLM-assisted urban simulation platform that scales to city-size populations while supporting empirical validation against real-world mobility patterns. Rather than invoking large language models for every agent action, it combines established human mobility models with fine-tuned cross-encoders to estimate semantic alignment between agent profiles, schedules, and activity transitions. A case study demonstrated simulation of 100,000 agents over 75 days in under one hour on a single consumer GPU.
No score is assigned. Sources and their independence are shown in the citation chain below.