How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns

Stephanie Brandl; Ruixiang Cui; Anders Søgaard

doi:10.18653/v1/2022.naacl-main.265

How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns

Stephanie Brandl, Ruixiang Cui, Anders Søgaard

Natural Language Processing

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › peer review

15 Citationer (Scopus)

31 Downloads (Pure)

Abstract

Gender-neutral pronouns have recently been introduced in many languages to a) include non-binary people and b) as a generic singular. Recent results from psycholinguistics suggest that gender-neutral pronouns (in Swedish) are not associated with human processing difficulties. This, we show, is in sharp contrast with automated processing. We show that gender-neutral pronouns in Danish, English, and Swedish are associated with higher perplexity, more dispersed attention patterns, and worse downstream performance. We argue that such conservativity in language models may limit widespread adoption of gender-neutral pronouns and must therefore be resolved.

Originalsprog	Engelsk
Titel	NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics : Human Language Technologies, Proceedings of the Conference
Forlag	Association for Computational Linguistics (ACL)
Publikationsdato	2022
Sider	3624-3630
ISBN (Elektronisk)	9781955917711
DOI	https://doi.org/10.18653/v1/2022.naacl-main.265
Status	Udgivet - 2022
Begivenhed	2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022 - Seattle, USA Varighed: 10 jul. 2022 → 15 jul. 2022

Konference

Konference	2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022
Land/Område	USA
By	Seattle
Periode	10/07/2022 → 15/07/2022
Sponsor	Amazon, Bloomberg, et al., Google Research, LIVE PERSON, Meta

Bibliografisk note

Funding Information:
This work was partially funded by the Platform Intelligence in News project, which is supported by Innovation Fund Denmark via the Grand Solutions program. We thank Vinit Ravishankar and Jonas Lotz for fruitful discussions and Daniel Her-shcovich and Yova Kementchedjhieva for proofreading and valuable inputs on the manuscript. We also thank Kellie Webster for her valuable input on gender-neutral pronouns.

Publisher Copyright:
© 2022 Association for Computational Linguistics.

Adgang til dokumentet

10.18653/v1/2022.naacl-main.265Licens: CC BY

FulltextForlagets udgivne version, 206 KBLicens: CC BY

Andre filer og links

Link to publication in Scopus

Citationsformater

Brandl, S., Cui, R., & Søgaard, A. (2022). How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns. I NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference (s. 3624-3630). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2022.naacl-main.265

How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns. / Brandl, Stephanie ; Cui, Ruixiang ; Søgaard, Anders.

NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2022. s. 3624-3630.

Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › peer review

Brandl, S , Cui, R & Søgaard, A 2022, How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns. i NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. Association for Computational Linguistics (ACL), s. 3624-3630, 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022, Seattle, USA, 10/07/2022. https://doi.org/10.18653/v1/2022.naacl-main.265

Brandl S , Cui R , Søgaard A. How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns. I NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. Association for Computational Linguistics (ACL). 2022. s. 3624-3630 https://doi.org/10.18653/v1/2022.naacl-main.265

Brandl, Stephanie ; Cui, Ruixiang ; Søgaard, Anders. / How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns. NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Proceedings of the Conference. Association for Computational Linguistics (ACL), 2022. s. 3624-3630

@inproceedings{4aa8878f9d9846cf8ace567389a589ca,

title = "How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns",

abstract = "Gender-neutral pronouns have recently been introduced in many languages to a) include non-binary people and b) as a generic singular. Recent results from psycholinguistics suggest that gender-neutral pronouns (in Swedish) are not associated with human processing difficulties. This, we show, is in sharp contrast with automated processing. We show that gender-neutral pronouns in Danish, English, and Swedish are associated with higher perplexity, more dispersed attention patterns, and worse downstream performance. We argue that such conservativity in language models may limit widespread adoption of gender-neutral pronouns and must therefore be resolved.",

author = "Stephanie Brandl and Ruixiang Cui and Anders S{\o}gaard",

note = "Publisher Copyright: {\textcopyright} 2022 Association for Computational Linguistics.; 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022 ; Conference date: 10-07-2022 Through 15-07-2022",

year = "2022",

doi = "10.18653/v1/2022.naacl-main.265",

language = "English",

pages = "3624--3630",

booktitle = "NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics",

publisher = "Association for Computational Linguistics (ACL)",

address = "United States",

}

TY - GEN

T1 - How Conservative are Language Models? Adapting to the Introduction of Gender-Neutral Pronouns

AU - Brandl, Stephanie

AU - Cui, Ruixiang

AU - Søgaard, Anders

PY - 2022

Y1 - 2022

N2 - Gender-neutral pronouns have recently been introduced in many languages to a) include non-binary people and b) as a generic singular. Recent results from psycholinguistics suggest that gender-neutral pronouns (in Swedish) are not associated with human processing difficulties. This, we show, is in sharp contrast with automated processing. We show that gender-neutral pronouns in Danish, English, and Swedish are associated with higher perplexity, more dispersed attention patterns, and worse downstream performance. We argue that such conservativity in language models may limit widespread adoption of gender-neutral pronouns and must therefore be resolved.

AB - Gender-neutral pronouns have recently been introduced in many languages to a) include non-binary people and b) as a generic singular. Recent results from psycholinguistics suggest that gender-neutral pronouns (in Swedish) are not associated with human processing difficulties. This, we show, is in sharp contrast with automated processing. We show that gender-neutral pronouns in Danish, English, and Swedish are associated with higher perplexity, more dispersed attention patterns, and worse downstream performance. We argue that such conservativity in language models may limit widespread adoption of gender-neutral pronouns and must therefore be resolved.

UR - http://www.scopus.com/inward/record.url?scp=85138426739&partnerID=8YFLogxK

U2 - 10.18653/v1/2022.naacl-main.265

DO - 10.18653/v1/2022.naacl-main.265

M3 - Article in proceedings

AN - SCOPUS:85138426739

SP - 3624

EP - 3630

BT - NAACL 2022 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics

PB - Association for Computational Linguistics (ACL)

T2 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022

Y2 - 10 July 2022 through 15 July 2022

ER -