Evaluating Gender Bias of Pre-trained Language Models in Natural Language Inference by Considering All Labels

arXiv:2309.09697v2 Announce Type: replace
Abstract: Discriminatory gender biases have been found in Pre-trained Language Models (PLMs) for multiple languages. In Natural Language Inference (NLI), existing bias evaluation methods have focused on the prediction results of a specific label out of three labels, such as neutral. However, such evaluation methods can be inaccurate since unique biased inferences are associated with unique prediction labels. Addressing this limitation, we propose a bias evaluation method for PLMs that considers all the three labels of …

abstract arxiv bias biases cs.cl evaluation found gender gender bias inference labels language language models languages multiple natural natural language prediction type