2024 ECCV ECCV 2024

Learning Scalable Model Soup on a Single GPU: An Efficient Subspace Training Strategy