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

Gender bias is a significant issue in machine translation, leading to ongoingresearch efforts in developing bias mitigation techniques. However, most worksfocus on debiasing bilingual models without much consideration for multilingualsystems. In this paper, we specifically target the gender bias issue ofmultilingual machine translation models for unambiguous cases where there is asingle correct translation, and propose a bias mitigation method based on anovel approach. Specifically, we propose Gender-Aware Contrastive Learning,GACL, which encodes contextual gender information into the representations ofnon-explicit gender words. Our method is target language-agnostic and isapplicable to pre-trained multilingual machine translation models viafine-tuning. Through multilingual evaluation, we show that our approachimproves gender accuracy by a wide margin without hampering translationperformance. We also observe that incorporated gender information transfers andbenefits other target languages regarding gender accuracy. Finally, wedemonstrate that our method is applicable and beneficial to models of varioussizes.

Target-Agnostic Gender-Aware Contrastive Learning for Mitigating Bias in Multilingual Machine Translation