@inproceedings{lu-etal-2023-reducing,
title = "Reducing Gender Bias in {NMT} with {FUDGE}",
author = {Lu, Tianshuai and
Aepli, No{\"e}mi and
Rios, Annette},
editor = "Vanmassenhove, Eva and
Savoldi, Beatrice and
Bentivogli, Luisa and
Daems, Joke and
Hackenbuchner, Jani{\c{c}}a",
booktitle = "Proceedings of the First Workshop on Gender-Inclusive Translation Technologies",
month = jun,
year = "2023",
address = "Tampere, Finland",
publisher = "European Association for Machine Translation",
url = "https://aclanthology.org/2023.gitt-1.6",
pages = "61--69",
abstract = "Gender bias appears in many neural machine translation (NMT) models and commercial translation software. Research has become more aware of this problem in recent years and there has been work on mitigating gender bias. However, the challenge of addressing gender bias in NMT persists. This work utilizes a controlled text generation method, Future Discriminators for Generation (FUDGE), to reduce the so-called Speaking As gender bias. This bias emerges when translating from English to a language that openly marks the gender of the speaker. We evaluate the model on MuST-SHE, a challenge set to specifically evaluate gender translation. The results demonstrate improvements in the translation accuracy of the feminine terms.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lu-etal-2023-reducing">
<titleInfo>
<title>Reducing Gender Bias in NMT with FUDGE</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tianshuai</namePart>
<namePart type="family">Lu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Noëmi</namePart>
<namePart type="family">Aepli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Annette</namePart>
<namePart type="family">Rios</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First Workshop on Gender-Inclusive Translation Technologies</title>
</titleInfo>
<name type="personal">
<namePart type="given">Eva</namePart>
<namePart type="family">Vanmassenhove</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Beatrice</namePart>
<namePart type="family">Savoldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luisa</namePart>
<namePart type="family">Bentivogli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joke</namePart>
<namePart type="family">Daems</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Janiça</namePart>
<namePart type="family">Hackenbuchner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Association for Machine Translation</publisher>
<place>
<placeTerm type="text">Tampere, Finland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Gender bias appears in many neural machine translation (NMT) models and commercial translation software. Research has become more aware of this problem in recent years and there has been work on mitigating gender bias. However, the challenge of addressing gender bias in NMT persists. This work utilizes a controlled text generation method, Future Discriminators for Generation (FUDGE), to reduce the so-called Speaking As gender bias. This bias emerges when translating from English to a language that openly marks the gender of the speaker. We evaluate the model on MuST-SHE, a challenge set to specifically evaluate gender translation. The results demonstrate improvements in the translation accuracy of the feminine terms.</abstract>
<identifier type="citekey">lu-etal-2023-reducing</identifier>
<location>
<url>https://aclanthology.org/2023.gitt-1.6</url>
</location>
<part>
<date>2023-06</date>
<extent unit="page">
<start>61</start>
<end>69</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Reducing Gender Bias in NMT with FUDGE
%A Lu, Tianshuai
%A Aepli, Noëmi
%A Rios, Annette
%Y Vanmassenhove, Eva
%Y Savoldi, Beatrice
%Y Bentivogli, Luisa
%Y Daems, Joke
%Y Hackenbuchner, Janiça
%S Proceedings of the First Workshop on Gender-Inclusive Translation Technologies
%D 2023
%8 June
%I European Association for Machine Translation
%C Tampere, Finland
%F lu-etal-2023-reducing
%X Gender bias appears in many neural machine translation (NMT) models and commercial translation software. Research has become more aware of this problem in recent years and there has been work on mitigating gender bias. However, the challenge of addressing gender bias in NMT persists. This work utilizes a controlled text generation method, Future Discriminators for Generation (FUDGE), to reduce the so-called Speaking As gender bias. This bias emerges when translating from English to a language that openly marks the gender of the speaker. We evaluate the model on MuST-SHE, a challenge set to specifically evaluate gender translation. The results demonstrate improvements in the translation accuracy of the feminine terms.
%U https://aclanthology.org/2023.gitt-1.6
%P 61-69
Markdown (Informal)
[Reducing Gender Bias in NMT with FUDGE](https://aclanthology.org/2023.gitt-1.6) (Lu et al., GITT 2023)
ACL
- Tianshuai Lu, Noëmi Aepli, and Annette Rios. 2023. Reducing Gender Bias in NMT with FUDGE. In Proceedings of the First Workshop on Gender-Inclusive Translation Technologies, pages 61–69, Tampere, Finland. European Association for Machine Translation.