Fusing Urban Structure and Semantics: A Conditional Diffusion Model for Cross-City OD Matrix Generation

SEDAN, a Structure-Enhanced Diffusion model for Origin-Destination matrix generation, frames cities as attributed graphs where regions are nodes carrying demographic and POI features, with commuting flows as weighted edges conditioned on adjacency and distance matrices. The model fuses spatial structure with semantic attributes via a conditional diffusion process, enabling cross-city generalization without retraining. Urban planners, traffic engineers, and smart city ML teams should note this directly tackles OD heterogeneity that cripples traditional gravity and neural flow models across unseen cities.