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

There are 293 scripts representing over 7,000 languages in the written form.Due to various reasons, many closely related languages use different scripts,which poses difficulty for multilingual pretrained language models (mPLMs) inlearning crosslingual knowledge through lexical overlap. As a result, mPLMspresent a script barrier: representations from different scripts are located indifferent subspaces, which is a strong indicator of why crosslingual transferinvolving languages of different scripts shows sub-optimal performance. Toaddress this problem, we propose a simple framework TransliCo that containsTransliteration Contrastive Modeling (TCM) to fine-tune an mPLM by contrastingsentences in its training data and their transliterations in a unified script(Latn, in our case), which ensures uniformity in the representation space fordifferent scripts. Using Glot500-m, an mPLM pretrained on over 500 languages,as our source model, we find-tune it on a small portion (5\%) of its trainingdata, and refer to the resulting model as Furina. We show that Furina not onlybetter aligns representations from distinct scripts but also outperforms theoriginal Glot500-m on various crosslingual transfer tasks. Additionally, weachieve consistent improvement in a case study on the Indic group where thelanguages are highly related but use different scripts. We make our code andmodels publicly available.

TransliCo: A Contrastive Learning Framework to Address the Script Barrier in Multilingual Pretrained Language Models