No idea. I think the process is pretty similar, isn't it? Or do ICML reviewers get fewer papers? Then they are double blind.