Evaluating Bias and Fairness in Gender-Neutral Pretrained Vision-and-Language Models
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Evaluating Bias and Fairness in Gender-Neutral Pretrained Vision-and-Language Models. / Cabello, Laura; Bugliarello, Emanuele; Brandl, Stephanie; Elliott, Desmond.
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics (ACL), 2023. p. 8465-8483.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Evaluating Bias and Fairness in Gender-Neutral Pretrained Vision-and-Language Models
AU - Cabello, Laura
AU - Bugliarello, Emanuele
AU - Brandl, Stephanie
AU - Elliott, Desmond
PY - 2023
Y1 - 2023
N2 - Pretrained machine learning models are known to perpetuate and even amplify existing biases in data, which can result in unfair outcomes that ultimately impact user experience. Therefore, it is crucial to understand the mechanisms behind those prejudicial biases to ensure that model performance does not result in discriminatory behaviour toward certain groups or populations. In this work, we define gender bias as our case study. We quantify bias amplification in pretraining and after fine-tuning on three families of vision-and-language models. We investigate the connection, if any, between the two learning stages, and evaluate how bias amplification reflects on model performance. Overall, we find that bias amplification in pretraining and after fine-tuning are independent. We then examine the effect of continued pretraining on gender-neutral data, finding that this reduces group disparities, i.e., promotes fairness, on VQAv2 and retrieval tasks without significantly compromising task performance.
AB - Pretrained machine learning models are known to perpetuate and even amplify existing biases in data, which can result in unfair outcomes that ultimately impact user experience. Therefore, it is crucial to understand the mechanisms behind those prejudicial biases to ensure that model performance does not result in discriminatory behaviour toward certain groups or populations. In this work, we define gender bias as our case study. We quantify bias amplification in pretraining and after fine-tuning on three families of vision-and-language models. We investigate the connection, if any, between the two learning stages, and evaluate how bias amplification reflects on model performance. Overall, we find that bias amplification in pretraining and after fine-tuning are independent. We then examine the effect of continued pretraining on gender-neutral data, finding that this reduces group disparities, i.e., promotes fairness, on VQAv2 and retrieval tasks without significantly compromising task performance.
KW - cs.CV
KW - cs.CL
KW - cs.LG
U2 - 10.18653/v1/2023.emnlp-main.525
DO - 10.18653/v1/2023.emnlp-main.525
M3 - Article in proceedings
SP - 8465
EP - 8483
BT - Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
PB - Association for Computational Linguistics (ACL)
T2 - 2023 Conference on Empirical Methods in Natural Language Processing
Y2 - 6 December 2023 through 10 December 2023
ER -
ID: 382997067