Synergizing Transport-Based Generative Models and Latent Geometry for Stochastic Closure Modeling

arXiv:2602.17089v1 Announce Type: cross Abstract: Diffusion models recently developed for generative AI tasks can produce high-quality samples while still maintaining diversity among samples to promote mode coverage, providing a promising path for learning stochastic closure m...

🔗 Read more: https://arxiv.org/abs/2602.17089

#News #Software #Energy #Policy #AI #Biology #Academic