aiqbench analysis
CityReal introduces a modular framework where large language model agents function as intention-driven decision-makers rather than relying on static few-shot prompting, significantly improving alignment with real-world urban population statistics. By enabling these agents to learn habits and adapt over time through textual adapters, the system supports scalable simulations of crowd dynamics, mobility flows, and well-being across tens of thousands of entities. This advancement offers a robust testbed for social science research, traffic safety analysis, and transportation policy forecasting without succumbing to model behavioral priors.
CityReal: Human-Aligned Urban Behavior and City Dynamics Simulation with Large-Scale LLM Agents
arXiv:2608.16897v1 Announce Type: cross Abstract: Large-scale urban simulation plays a pivotal role in social science, traffic safety, and transportation policy. Recent work has shown that large language models, when prompted as agents, can generate lifelike daily routines at city scale. Yet these methods typically rely on few-shot prompting, causing agents to reproduce the LLM's behavioral priors rather than the target population. We introduce CityReal, a modular framework for human-aligned urban simulation. CityReal models agents as intention-driven decision makers that pursue coherent mobility and activity plans rather than isolated step-by-step choices. They adapt over time by learning habits and preferences based on experience and constraints. To improve population-level realism, we learn textual adapters for behavior modules that align agent decisions with observed population statistics. Experiments show that CityReal improves alignment with real-world human behavior at both micro and macro levels. Scaling to tens of thousands of agents, it supports analysis of crowd density, place popularity, mobility flows, and well-being under different urban scenarios, offering a scalable testbed for urban simulation and forecasting.
Abstract:Large-scale urban simulation plays a pivotal role in social science, traffic safety, and transportation policy. Recent work has shown that large language models, when prompted as agents, can generate lifelike daily routines at city scale. Yet these methods typically rely on few-shot prompting, causing agents to reproduce the LLM's behavioral priors rather than the target population. We introduce CityReal, a modular framework for human-aligned urban simulation. CityReal models agents as intention-driven decision makers that pursue coherent mobility and activity plans rather than isolated step-by-step choices. They adapt over time by learning habits and preferences based on experience and constraints. To improve population-level realism, we learn textual adapters for behavior modules that align agent decisions with observed population statistics. Experiments show that CityReal improves alignment with real-world human behavior at both micro and macro levels. Scaling to tens of thousands of agents, it supports analysis of crowd density, place popularity, mobility flows, and well-being under different urban scenarios, offering a scalable testbed for urban simulation and forecasting.
Physics and Society (physics.soc-ph); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
arXiv:2608.16897 [physics.soc-ph]
(or arXiv:2608.16897v1 [physics.soc-ph] for this version)
https://doi.org/10.48550/arXiv.2608.16897 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 265 words
