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

The vast majority of today's large language models are English-centric,having been pretrained predominantly on English text. Yet, in order to meetuser expectations, models need to be able to respond appropriately in multiplelanguages once deployed in downstream applications. Given limited exposure toother languages during pretraining, cross-lingual transfer is important forachieving decent performance in non-English settings. In this work, weinvestigate just how much multilinguality is required during finetuning toelicit strong cross-lingual generalisation across a range of tasks and targetlanguages. We find that, compared to English-only finetuning, multilingualinstruction tuning with as few as three languages significantly improves amodel's cross-lingual transfer abilities on generative tasks that assumeinput/output language agreement, while being of less importance for highlystructured tasks. Our code and data is available athttps://github.com/ZurichNLP/multilingual-instruction-tuning.

Turning English-centric LLMs Into Polyglots: How Much Multilinguality Is Needed?