Sociodemographic Bias in Language Models: A Survey and Forward Path

Vipul Gupta; Pranav Narayanan Venkit; Shomir Wilson; Rebecca J. Passonneau

Sociodemographic Bias in Language Models: A Survey and Forward Path

Vipul Gupta, Pranav Narayanan Venkit, Shomir Wilson, Rebecca J. Passonneau

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Sociodemographic bias in language models (LMs) has the potential for harm when deployed in real-world settings. This paper presents a comprehensive survey of the past decade of research on sociodemographic bias in LMs, organized into a typology that facilitates examining the different aims: types of bias, quantifying bias, and debiasing techniques. We track the evolution of the latter two questions, then identify current trends and their limitations, as well as emerging techniques. To guide future research towards more effective and reliable solutions, and to help authors situate their work within this broad landscape, we conclude with a checklist of open questions.

Original language	English (US)
Title of host publication	GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop
Editors	Agnieszka Falenska, Christine Basta, Marta Costa-jussa, Seraphina Goldfarb-Tarrant, Debora Nozza
Publisher	Association for Computational Linguistics (ACL)
Pages	295-322
Number of pages	28
ISBN (Electronic)	9798891761377
State	Published - 2024
Event	5th Workshop on Gender Bias in Natural Language Processing, GeBNLP 2024, held in conjunction with the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024 - Bangkok, Thailand Duration: Aug 16 2024 → …

Publication series

Name	GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop

Conference

Conference	5th Workshop on Gender Bias in Natural Language Processing, GeBNLP 2024, held in conjunction with the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024
Country/Territory	Thailand
City	Bangkok
Period	8/16/24 → …

All Science Journal Classification (ASJC) codes

Computational Theory and Mathematics
Computer Science Applications
Information Systems
General Psychology
Gender Studies

Cite this

Gupta, V., Venkit, P. N., Wilson, S., & Passonneau, R. J. (2024). Sociodemographic Bias in Language Models: A Survey and Forward Path. In A. Falenska, C. Basta, M. Costa-jussa, S. Goldfarb-Tarrant, & D. Nozza (Eds.), GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop (pp. 295-322). (GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop). Association for Computational Linguistics (ACL).

Gupta, Vipul ; Venkit, Pranav Narayanan ; Wilson, Shomir et al. / Sociodemographic Bias in Language Models : A Survey and Forward Path. GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop. editor / Agnieszka Falenska ; Christine Basta ; Marta Costa-jussa ; Seraphina Goldfarb-Tarrant ; Debora Nozza. Association for Computational Linguistics (ACL), 2024. pp. 295-322 (GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop).

@inproceedings{d88b2a6f4443475ebb53331e53d8a3d9,

title = "Sociodemographic Bias in Language Models: A Survey and Forward Path",

abstract = "Sociodemographic bias in language models (LMs) has the potential for harm when deployed in real-world settings. This paper presents a comprehensive survey of the past decade of research on sociodemographic bias in LMs, organized into a typology that facilitates examining the different aims: types of bias, quantifying bias, and debiasing techniques. We track the evolution of the latter two questions, then identify current trends and their limitations, as well as emerging techniques. To guide future research towards more effective and reliable solutions, and to help authors situate their work within this broad landscape, we conclude with a checklist of open questions.",

author = "Vipul Gupta and Venkit, {Pranav Narayanan} and Shomir Wilson and Passonneau, {Rebecca J.}",

note = "Publisher Copyright: {\textcopyright}2024 Association for Computational Linguistics.; 5th Workshop on Gender Bias in Natural Language Processing, GeBNLP 2024, held in conjunction with the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024 ; Conference date: 16-08-2024",

year = "2024",

language = "English (US)",

series = "GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop",

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

pages = "295--322",

editor = "Agnieszka Falenska and Christine Basta and Marta Costa-jussa and Seraphina Goldfarb-Tarrant and Debora Nozza",

booktitle = "GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop",

}

Gupta, V, Venkit, PN, Wilson, S & Passonneau, RJ 2024, Sociodemographic Bias in Language Models: A Survey and Forward Path. in A Falenska, C Basta, M Costa-jussa, S Goldfarb-Tarrant & D Nozza (eds), GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop. GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop, Association for Computational Linguistics (ACL), pp. 295-322, 5th Workshop on Gender Bias in Natural Language Processing, GeBNLP 2024, held in conjunction with the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024, Bangkok, Thailand, 8/16/24.

Sociodemographic Bias in Language Models: A Survey and Forward Path. / Gupta, Vipul; Venkit, Pranav Narayanan; Wilson, Shomir et al.
GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop. ed. / Agnieszka Falenska; Christine Basta; Marta Costa-jussa; Seraphina Goldfarb-Tarrant; Debora Nozza. Association for Computational Linguistics (ACL), 2024. p. 295-322 (GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Sociodemographic Bias in Language Models

T2 - 5th Workshop on Gender Bias in Natural Language Processing, GeBNLP 2024, held in conjunction with the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024

AU - Gupta, Vipul

AU - Venkit, Pranav Narayanan

AU - Wilson, Shomir

AU - Passonneau, Rebecca J.

PY - 2024

Y1 - 2024

N2 - Sociodemographic bias in language models (LMs) has the potential for harm when deployed in real-world settings. This paper presents a comprehensive survey of the past decade of research on sociodemographic bias in LMs, organized into a typology that facilitates examining the different aims: types of bias, quantifying bias, and debiasing techniques. We track the evolution of the latter two questions, then identify current trends and their limitations, as well as emerging techniques. To guide future research towards more effective and reliable solutions, and to help authors situate their work within this broad landscape, we conclude with a checklist of open questions.

AB - Sociodemographic bias in language models (LMs) has the potential for harm when deployed in real-world settings. This paper presents a comprehensive survey of the past decade of research on sociodemographic bias in LMs, organized into a typology that facilitates examining the different aims: types of bias, quantifying bias, and debiasing techniques. We track the evolution of the latter two questions, then identify current trends and their limitations, as well as emerging techniques. To guide future research towards more effective and reliable solutions, and to help authors situate their work within this broad landscape, we conclude with a checklist of open questions.

UR - http://www.scopus.com/inward/record.url?scp=85204371955&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85204371955&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85204371955

T3 - GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop

SP - 295

EP - 322

BT - GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop

A2 - Falenska, Agnieszka

A2 - Basta, Christine

A2 - Costa-jussa, Marta

A2 - Goldfarb-Tarrant, Seraphina

A2 - Nozza, Debora

PB - Association for Computational Linguistics (ACL)

Y2 - 16 August 2024

ER -

Gupta V, Venkit PN, Wilson S , Passonneau RJ. Sociodemographic Bias in Language Models: A Survey and Forward Path. In Falenska A, Basta C, Costa-jussa M, Goldfarb-Tarrant S, Nozza D, editors, GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop. Association for Computational Linguistics (ACL). 2024. p. 295-322. (GeBNLP 2024 - 5th Workshop on Gender Bias in Natural Language Processing, Proceedings of the Workshop).

Sociodemographic Bias in Language Models: A Survey and Forward Path

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this