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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Self-supervised Learning (SSL) including the mainstream contrastive learninghas achieved great success in learning visual representations without dataannotations. However, most of methods mainly focus on the instance levelinformation (\ie, the different augmented images of the same instance shouldhave the same feature or cluster into the same class), but there is a lack ofattention on the relationships between different instances. In this paper, weintroduced a novel SSL paradigm, which we term as relational self-supervisedlearning (ReSSL) framework that learns representations by modeling therelationship between different instances. Specifically, our proposed methodemploys sharpened distribution of pairwise similarities among differentinstances as \textit{relation} metric, which is thus utilized to match thefeature embeddings of different augmentations. Moreover, to boost theperformance, we argue that weak augmentations matter to represent a morereliable relation, and leverage momentum strategy for practical efficiency.Experimental results show that our proposed ReSSL significantly outperforms theprevious state-of-the-art algorithms in terms of both performance and trainingefficiency. Code is available at \url{https://github.com/KyleZheng1997/ReSSL}.

ReSSL: Relational Self-Supervised Learning with Weak Augmentation