<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sedoc-ungar-2019-role">
    <titleInfo>
        <title>The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">João</namePart>
        <namePart type="family">Sedoc</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Lyle</namePart>
        <namePart type="family">Ungar</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2019-08</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the First Workshop on Gender Bias in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Marta</namePart>
            <namePart type="given">R</namePart>
            <namePart type="family">Costa-jussà</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Christian</namePart>
            <namePart type="family">Hardmeier</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Will</namePart>
            <namePart type="family">Radford</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kellie</namePart>
            <namePart type="family">Webster</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Florence, Italy</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Systemic bias in word embeddings has been widely reported and studied, and efforts made to debias them; however, new contextualized embeddings such as ELMo and BERT are only now being similarly studied. Standard debiasing methods require heterogeneous lists of target words to identify the “bias subspace”. We show show that using new contextualized word embeddings in conceptor debiasing allows us to more accurately debias word embeddings by breaking target word lists into more homogeneous subsets and then combining (”Or’ing”) the debiasing conceptors of the different subsets.</abstract>
    <identifier type="citekey">sedoc-ungar-2019-role</identifier>
    <identifier type="doi">10.18653/v1/W19-3808</identifier>
    <location>
        <url>https://aclanthology.org/W19-3808</url>
    </location>
    <part>
        <date>2019-08</date>
        <extent unit="page">
            <start>55</start>
            <end>61</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations
%A Sedoc, João
%A Ungar, Lyle
%Y Costa-jussà, Marta R.
%Y Hardmeier, Christian
%Y Radford, Will
%Y Webster, Kellie
%S Proceedings of the First Workshop on Gender Bias in Natural Language Processing
%D 2019
%8 August
%I Association for Computational Linguistics
%C Florence, Italy
%F sedoc-ungar-2019-role
%X Systemic bias in word embeddings has been widely reported and studied, and efforts made to debias them; however, new contextualized embeddings such as ELMo and BERT are only now being similarly studied. Standard debiasing methods require heterogeneous lists of target words to identify the “bias subspace”. We show show that using new contextualized word embeddings in conceptor debiasing allows us to more accurately debias word embeddings by breaking target word lists into more homogeneous subsets and then combining (”Or’ing”) the debiasing conceptors of the different subsets.
%R 10.18653/v1/W19-3808
%U https://aclanthology.org/W19-3808
%U https://doi.org/10.18653/v1/W19-3808
%P 55-61

Markdown (Informal)

[The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations](https://aclanthology.org/W19-3808) (Sedoc & Ungar, GeBNLP 2019)

• The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations (Sedoc & Ungar, GeBNLP 2019)

ACL

• João Sedoc and Lyle Ungar. 2019. The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations. In Proceedings of the First Workshop on Gender Bias in Natural Language Processing, pages 55–61, Florence, Italy. Association for Computational Linguistics.