@inproceedings{gonzalez-etal-2020-type,
title = "Type {B} Reflexivization as an Unambiguous Testbed for Multilingual Multi-Task Gender Bias",
author = "Gonz{\'a}lez, Ana Valeria and
Barrett, Maria and
Hvingelby, Rasmus and
Webster, Kellie and
S{\o}gaard, Anders",
editor = "Webber, Bonnie and
Cohn, Trevor and
He, Yulan and
Liu, Yang",
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.emnlp-main.209/",
doi = "10.18653/v1/2020.emnlp-main.209",
pages = "2637--2648",
abstract = "The one-sided focus on English in previous studies of gender bias in NLP misses out on opportunities in other languages: English challenge datasets such as GAP and WinoGender highlight model preferences that are {\textquotedblleft}hallucinatory{\textquotedblright}, e.g., disambiguating gender-ambiguous occurrences of {\textquoteleft}doctor' as male doctors. We show that for languages with type B reflexivization, e.g., Swedish and Russian, we can construct multi-task challenge datasets for detecting gender bias that lead to unambiguously wrong model predictions: In these languages, the direct translation of {\textquoteleft}the doctor removed his mask' is not ambiguous between a coreferential reading and a disjoint reading. Instead, the coreferential reading requires a non-gendered pronoun, and the gendered, possessive pronouns are anti-reflexive. We present a multilingual, multi-task challenge dataset, which spans four languages and four NLP tasks and focuses only on this phenomenon. We find evidence for gender bias across all task-language combinations and correlate model bias with national labor market statistics."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gonzalez-etal-2020-type">
<titleInfo>
<title>Type B Reflexivization as an Unambiguous Testbed for Multilingual Multi-Task Gender Bias</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ana</namePart>
<namePart type="given">Valeria</namePart>
<namePart type="family">González</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Barrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rasmus</namePart>
<namePart type="family">Hvingelby</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kellie</namePart>
<namePart type="family">Webster</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anders</namePart>
<namePart type="family">Søgaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bonnie</namePart>
<namePart type="family">Webber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Trevor</namePart>
<namePart type="family">Cohn</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yulan</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The one-sided focus on English in previous studies of gender bias in NLP misses out on opportunities in other languages: English challenge datasets such as GAP and WinoGender highlight model preferences that are “hallucinatory”, e.g., disambiguating gender-ambiguous occurrences of ‘doctor’ as male doctors. We show that for languages with type B reflexivization, e.g., Swedish and Russian, we can construct multi-task challenge datasets for detecting gender bias that lead to unambiguously wrong model predictions: In these languages, the direct translation of ‘the doctor removed his mask’ is not ambiguous between a coreferential reading and a disjoint reading. Instead, the coreferential reading requires a non-gendered pronoun, and the gendered, possessive pronouns are anti-reflexive. We present a multilingual, multi-task challenge dataset, which spans four languages and four NLP tasks and focuses only on this phenomenon. We find evidence for gender bias across all task-language combinations and correlate model bias with national labor market statistics.</abstract>
<identifier type="citekey">gonzalez-etal-2020-type</identifier>
<identifier type="doi">10.18653/v1/2020.emnlp-main.209</identifier>
<location>
<url>https://aclanthology.org/2020.emnlp-main.209/</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>2637</start>
<end>2648</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Type B Reflexivization as an Unambiguous Testbed for Multilingual Multi-Task Gender Bias
%A González, Ana Valeria
%A Barrett, Maria
%A Hvingelby, Rasmus
%A Webster, Kellie
%A Søgaard, Anders
%Y Webber, Bonnie
%Y Cohn, Trevor
%Y He, Yulan
%Y Liu, Yang
%S Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F gonzalez-etal-2020-type
%X The one-sided focus on English in previous studies of gender bias in NLP misses out on opportunities in other languages: English challenge datasets such as GAP and WinoGender highlight model preferences that are “hallucinatory”, e.g., disambiguating gender-ambiguous occurrences of ‘doctor’ as male doctors. We show that for languages with type B reflexivization, e.g., Swedish and Russian, we can construct multi-task challenge datasets for detecting gender bias that lead to unambiguously wrong model predictions: In these languages, the direct translation of ‘the doctor removed his mask’ is not ambiguous between a coreferential reading and a disjoint reading. Instead, the coreferential reading requires a non-gendered pronoun, and the gendered, possessive pronouns are anti-reflexive. We present a multilingual, multi-task challenge dataset, which spans four languages and four NLP tasks and focuses only on this phenomenon. We find evidence for gender bias across all task-language combinations and correlate model bias with national labor market statistics.
%R 10.18653/v1/2020.emnlp-main.209
%U https://aclanthology.org/2020.emnlp-main.209/
%U https://doi.org/10.18653/v1/2020.emnlp-main.209
%P 2637-2648
Markdown (Informal)
[Type B Reflexivization as an Unambiguous Testbed for Multilingual Multi-Task Gender Bias](https://aclanthology.org/2020.emnlp-main.209/) (González et al., EMNLP 2020)
ACL