Gender-Neutral Machine Translation Strategies in Practice
This addresses the problem of misgendering and representational harms in machine translation for users of gender-inclusive language, but it is incremental as it evaluates existing systems rather than proposing new methods.
The study assessed 21 machine translation systems for their ability to maintain gender neutrality when translating ambiguous gender from English to grammatical gender languages, finding a general lack of gender-neutral translations but noting a few systems that used specific strategies depending on the target language.
Gender-inclusive machine translation (MT) should preserve gender ambiguity in the source to avoid misgendering and representational harms. While gender ambiguity often occurs naturally in notional gender languages such as English, maintaining that gender neutrality in grammatical gender languages is a challenge. Here we assess the sensitivity of 21 MT systems to the need for gender neutrality in response to gender ambiguity in three translation directions of varying difficulty. The specific gender-neutral strategies that are observed in practice are categorized and discussed. Additionally, we examine the effect of binary gender stereotypes on the use of gender-neutral translation. In general, we report a disappointing absence of gender-neutral translations in response to gender ambiguity. However, we observe a small handful of MT systems that switch to gender neutral translation using specific strategies, depending on the target language.