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

Recent impressive improvements in NLP, largely based on the success ofcontextual neural language models, have been mostly demonstrated on at most acouple dozen high-resource languages. Building language models and, moregenerally, NLP systems for non-standardized and low-resource languages remainsa challenging task. In this work, we focus on North-African colloquialdialectal Arabic written using an extension of the Latin script, calledNArabizi, found mostly on social media and messaging communication. In thislow-resource scenario with data displaying a high level of variability, wecompare the downstream performance of a character-based language model onpart-of-speech tagging and dependency parsing to that of monolingual andmultilingual models. We show that a character-based model trained on only 99ksentences of NArabizi and fined-tuned on a small treebank of this languageleads to performance close to those obtained with the same architecturepre-trained on large multilingual and monolingual models. Confirming theseresults a on much larger data set of noisy French user-generated content, weargue that such character-based language models can be an asset for NLP inlow-resource and high language variability set-tings.

Can Character-based Language Models Improve Downstream Task Performance in Low-Resource and Noisy Language Scenarios?