MoSA: Mixture of Sparse Adapters for Visual Efficient Tuning

arXiv:2312.02923v2 Announce Type: replace
Abstract: With the rapid growth in the scale of pre-trained foundation models, parameter-efficient fine-tuning techniques have gained significant attention, among which Adapter Tuning is the most widely used. Despite achieving efficiency, it still underperforms full fine-tuning, and the performance improves at the cost of an increase in parameters. Recent efforts have either focused on training multiple adapter experts to increase model capacity or on pruning adapters to achieve parameter efficiency. However, both approaches introduce more parameters …

abstract adapter arxiv attention cost cs.cv efficiency fine-tuning foundation growth performance scale type visual