Counterfactually Augmented Data and Unintended Bias: The Case of Sexism and Hate Speech Detection
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Counterfactually Augmented Data and Unintended Bias : The Case of Sexism and Hate Speech Detection. / Sen, Indira; Samory, Mattia; Wagner, Claudia; Augenstein, Isabelle.
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Association for Computational Linguistics (ACL), 2022. p. 4716-4726.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Counterfactually Augmented Data and Unintended Bias
T2 - 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL 2022
AU - Sen, Indira
AU - Samory, Mattia
AU - Wagner, Claudia
AU - Augenstein, Isabelle
N1 - Publisher Copyright: © 2022 Association for Computational Linguistics.
PY - 2022
Y1 - 2022
N2 - Counterfactually Augmented Data (CAD) aims to improve out-of-domain generalizability, an indicator of model robustness. The improvement is credited to promoting core features of the construct over spurious artifacts that happen to correlate with it. Yet, over-relying on core features may lead to unintended model bias. Especially, construct-driven CAD-perturbations of core features-may induce models to ignore the context in which core features are used. Here, we test models for sexism and hate speech detection on challenging data: non-hateful and nonsexist usage of identity and gendered terms. On these hard cases, models trained on CAD, especially construct-driven CAD, show higher false positive rates than models trained on the original, unperturbed data. Using a diverse set of CAD-construct-driven and construct-agnostic-reduces such unintended bias.
AB - Counterfactually Augmented Data (CAD) aims to improve out-of-domain generalizability, an indicator of model robustness. The improvement is credited to promoting core features of the construct over spurious artifacts that happen to correlate with it. Yet, over-relying on core features may lead to unintended model bias. Especially, construct-driven CAD-perturbations of core features-may induce models to ignore the context in which core features are used. Here, we test models for sexism and hate speech detection on challenging data: non-hateful and nonsexist usage of identity and gendered terms. On these hard cases, models trained on CAD, especially construct-driven CAD, show higher false positive rates than models trained on the original, unperturbed data. Using a diverse set of CAD-construct-driven and construct-agnostic-reduces such unintended bias.
UR - http://www.scopus.com/inward/record.url?scp=85138414469&partnerID=8YFLogxK
U2 - 10.18653/v1/2022.naacl-main.347
DO - 10.18653/v1/2022.naacl-main.347
M3 - Article in proceedings
AN - SCOPUS:85138414469
SP - 4716
EP - 4726
BT - Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
PB - Association for Computational Linguistics (ACL)
Y2 - 10 July 2022 through 15 July 2022
ER -
ID: 341054609