Counter-GAP: Counterfactual Bias Evaluation through Gendered Ambiguous Pronouns
This work addresses gender bias evaluation in NLP for researchers and practitioners, offering a more reliable dataset and metric, though it is incremental in improving existing bias measurement methods.
The authors tackled the problem of evaluating gender bias in language models through coreference resolution by constructing Counter-GAP, a dataset of 4008 instances grouped into 1002 quadruples, and found that four pre-trained models showed significantly higher inconsistency across gender groups than within them, with name-based data augmentation being more effective for bias mitigation than anonymization.
Bias-measuring datasets play a critical role in detecting biased behavior of language models and in evaluating progress of bias mitigation methods. In this work, we focus on evaluating gender bias through coreference resolution, where previous datasets are either hand-crafted or fail to reliably measure an explicitly defined bias. To overcome these shortcomings, we propose a novel method to collect diverse, natural, and minimally distant text pairs via counterfactual generation, and construct Counter-GAP, an annotated dataset consisting of 4008 instances grouped into 1002 quadruples. We further identify a bias cancellation problem in previous group-level metrics on Counter-GAP, and propose to use the difference between inconsistency across genders and within genders to measure bias at a quadruple level. Our results show that four pre-trained language models are significantly more inconsistent across different gender groups than within each group, and that a name-based counterfactual data augmentation method is more effective to mitigate such bias than an anonymization-based method.