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

We propose a self-supervised shared encoder model that achieves strongresults on several visual, language and multimodal benchmarks while being data,memory and run-time efficient. We make three key contributions. First, incontrast to most existing works, we use a single transformer with all theencoder layers processing both the text and the image modalities. Second, wepropose a stage-wise training strategy where the model is first trained onimages, then jointly with unimodal text and image datasets and finally jointlywith text and text-image datasets. Third, to preserve information across boththe modalities, we propose a training pipeline that learns simultaneously fromgradient updates of different modalities at each training update step. Theresults on downstream text-only, image-only and multimodal tasks show that ourmodel is competitive with several strong models while using fewer parametersand lesser pre-training data. For example, MoMo performs competitively withFLAVA on multimodal (+3.1), image-only (+1.1) and text-only (-0.1) tasksdespite having 2/5th the number of parameters and using 1/3rd the image-texttraining pairs. Finally, we ablate various design choices and further show thatincreasing model size produces significant performance gains indicatingpotential for substantial improvements with larger models using our approach.

MoMo: A shared encoder Model for text, image and multi-Modal representations