Diffusion model fuses urban data for improved commuting flow prediction

Researchers have developed SEDAN, a novel conditional diffusion model designed to generate Origin-Destination (OD) matrices for commuting flows across different cities. This model represents cities as attributed graphs, incorporating demographic and point-of-interest features along with spatial structure like adjacency and distance matrices. SEDAN fuses semantic and spatial information to create more accurate and generalizable OD matrices, outperforming existing methods by 7.38% in RMSE on U.S. city datasets. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Provides a more accurate and generalizable method for urban planning and resource allocation by improving OD matrix generation.

RANK_REASON This is a research paper detailing a new model for OD matrix generation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• SEDAN
• arXiv
• U.S.

• paper
• other