A Study of Implicit Bias in Pretrained Language Models against People with Disabilities

Pranav Narayanan Venkit, Mukund Srinath, Shomir Wilson


Abstract
Pretrained language models (PLMs) have been shown to exhibit sociodemographic biases, such as against gender and race, raising concerns of downstream biases in language technologies. However, PLMs’ biases against people with disabilities (PWDs) have received little attention, in spite of their potential to cause similar harms. Using perturbation sensitivity analysis, we test an assortment of popular word embedding-based and transformer-based PLMs and show significant biases against PWDs in all of them. The results demonstrate how models trained on large corpora widely favor ableist language.
Anthology ID:
2022.coling-1.113
Volume:
Proceedings of the 29th International Conference on Computational Linguistics
Month:
October
Year:
2022
Address:
Gyeongju, Republic of Korea
Venue:
COLING
SIG:
Publisher:
International Committee on Computational Linguistics
Note:
Pages:
1324–1332
Language:
URL:
https://aclanthology.org/2022.coling-1.113
DOI:
Bibkey:
Cite (ACL):
Pranav Narayanan Venkit, Mukund Srinath, and Shomir Wilson. 2022. A Study of Implicit Bias in Pretrained Language Models against People with Disabilities. In Proceedings of the 29th International Conference on Computational Linguistics, pages 1324–1332, Gyeongju, Republic of Korea. International Committee on Computational Linguistics.
Cite (Informal):
A Study of Implicit Bias in Pretrained Language Models against People with Disabilities (Venkit et al., COLING 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.coling-1.113.pdf