Evaluation of adaptive machine translation from a gender-neutral language perspective

Publication type
C1
Publication status
Published
Authors
Kostikova, A., Lazarov, T., & Daems, J
Editor
Joss Moorkens and Vilelmini Sosoni
Series
44th Translating and the Computer conference (TC44), Proceedings
Pagination
82-89
Publisher
Tradulex (Geneva)
Conference
44th Translating and the Computer conference (TC44) (Luxembourg)
Download
(.pdf)
View in Biblio
(externe link)

Abstract

In this paper, we attempt to analyse the problem of conveying gender-neutral language when working with notional and grammatical languages (English and German) from the point of view of adaptive machine translation (MT). More specifically, we assess the efficiency of adaptive MT when it comes to gender-neutral language use, the purpose of which is to "reduce gender stereotyping, promote social change and contribute to achieving gender equality". We conclude that the initial output largely reflects cases of misgendering and generic masculine – problems that are well documented in the MT field, but which still remain unresolved. Moreover, our experiment revealed that ModernMT faces systematic difficulties in adapting to gender-neutral language when working with the English-German translation direction.