This project addresses the challenge of generating realistic synthetic runoff data for multi-station hydrological systems. The input is a historical runoff matrix of shape (238 stations, 365 days), ...
CGDiT (Crystal Graph Diffusion Transformer) is a graph-based diffusion model designed for crystal structure generation. It combines graph neural networks with continuous coordinate/lattice diffusion ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results