English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

While parameter efficient tuning (PET) methods have shown great potentialwith transformer architecture on Natural Language Processing (NLP) tasks, theireffectiveness is still under-studied with large-scale ConvNets on ComputerVision (CV) tasks. This paper proposes Conv-Adapter, a PET module designed forConvNets. Conv-Adapter is light-weight, domain-transferable, andarchitecture-agnostic with generalized performance on different tasks. Whentransferring on downstream tasks, Conv-Adapter learns tasks-specific featuremodulation to the intermediate representations of backbone while keeping thepre-trained parameters frozen. By introducing only a tiny amount of learnableparameters, e.g., only 3.5% full fine-tuning parameters of ResNet50,Conv-Adapter outperforms previous PET baseline methods and achieves comparableor surpasses the performance of full fine-tuning on 23 classification tasks ofvarious domains. It also presents superior performance on few-shotclassifications, with an average margin of 3.39%. Beyond classification,Conv-Adapter can generalize to detection and segmentation tasks with more than50% reduction of parameters but comparable performance to the traditional fullfine-tuning.

Conv-Adapter: Exploring Parameter Efficient Transfer Learning for ConvNets