UrbanWell: Benchmarking Multimodal Large Language Models for Spatio-Temporal Urban Wellbeing Analytics
Researchers have introduced UrbanWell, a new benchmark designed to evaluate the spatio-temporal reasoning capabilities of multimodal large language models (MLLMs) in urban wellbeing analytics. The benchmark integrates diverse data, including satellite and street view imagery, environmental conditions, spatial accessibility, urban form, vitality, and subjective perceptions across 38 cities over multiple years. UrbanWell also includes tasks for forecasting future values and classifying temporal trends, providing a comprehensive assessment of MLLMs' performance in understanding complex urban dynamics. AI
IMPACT Establishes a new standard for evaluating MLLMs in complex urban analytics, potentially driving advancements in AI's ability to understand and improve city environments.