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

Using a language model (LM) pretrained on two languages with largemonolingual data in order to initialize an unsupervised neural machinetranslation (UNMT) system yields state-of-the-art results. When limited data isavailable for one language, however, this method leads to poor translations. Wepresent an effective approach that reuses an LM that is pretrained only on thehigh-resource language. The monolingual LM is fine-tuned on both languages andis then used to initialize a UNMT model. To reuse the pretrained LM, we have tomodify its predefined vocabulary, to account for the new language. We thereforepropose a novel vocabulary extension method. Our approach, RE-LM, outperforms acompetitive cross-lingual pretraining model (XLM) in English-Macedonian (En-Mk)and English-Albanian (En-Sq), yielding more than +8.3 BLEU points for all fourtranslation directions.

Reusing a Pretrained Language Model on Languages with Limited Corpora for Unsupervised NMT