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

Supervised deep learning-based approaches have been applied to task-orienteddialog and have proven to be effective for limited domain and languageapplications when a sufficient number of training examples are available. Inpractice, these approaches suffer from the drawbacks of domain-driven designand under-resourced languages. Domain and language models are supposed to growand change as the problem space evolves. On one hand, research on transferlearning has demonstrated the cross-lingual ability of multilingualTransformers-based models to learn semantically rich representations. On theother, in addition to the above approaches, meta-learning have enabled thedevelopment of task and language learning algorithms capable of fargeneralization. Through this context, this article proposes to investigate thecross-lingual transferability of using synergistically few-shot learning withprototypical neural networks and multilingual Transformers-based models.Experiments in natural language understanding tasks on MultiATIS++ corpus showsthat our approach substantially improves the observed transfer learningperformances between the low and the high resource languages. More generallyour approach confirms that the meaningful latent space learned in a givenlanguage can be can be generalized to unseen and under-resourced ones usingmeta-learning.

On the cross-lingual transferability of multilingual prototypical models across NLU tasks