Reproduce “Large language model empowered participatory urban planning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhilun Zhou, Yuming Lin, Yong Li
Researchers from Tsinghua University developed an LLM-powered framework for participatory urban planning that simulates thousands of diverse residents to generate human-centric land-use plans. This approach significantly enhances resident satisfaction and inclusivity for vulnerable groups compared to traditional methods and DRL, while maintaining competitive objective metrics for service and ecology.
Key insights
- Large Language Models can effectively create and manage a large-scale multi-agent system (1000+ agents) capable of emulating diverse human stakeholders with distinct profiles and needs in a complex domain like urban planning.
- Collaborative generation and iterative feedback loops driven by LLM agents substantively integrate human-centric demands, enabling planning schemes to reflect nuanced resident preferences and improve inclusion for vulnerable groups.
- LLM-powered participatory simulation enhances the transparency and interpretability of planning rationales, allowing for dynamic adjustments that balance objective planning constraints with subjective human satisfaction.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.