Conv-Adapter: Exploring Parameter Efficient Transfer Learning for ConvNets

arXiv:2208.07463v4 Announce Type: replace
Abstract: While parameter efficient tuning (PET) methods have shown great potential with transformer architecture on Natural Language Processing (NLP) tasks, their effectiveness with large-scale ConvNets is still under-studied on Computer Vision (CV) tasks. This paper proposes Conv-Adapter, a PET module designed for ConvNets. Conv-Adapter is light-weight, domain-transferable, and architecture-agnostic with generalized performance on different tasks. When transferring on downstream tasks, Conv-Adapter learns tasks-specific feature modulation to the intermediate representations of backbones while keeping the pre-trained parameters …

abstract adapter architecture arxiv computer computer vision cs.ai cs.cv domain language language processing light natural natural language natural language processing nlp paper pet processing scale tasks transfer transfer learning transformer transformer architecture type vision