Gender Bias in BERT -- Measuring and Analysing Biases through Sentiment Rating in a Realistic Downstream Classification Task

Abstract

Pretrained language models are publicly available and constantly finetuned for various real-life applications. As they become capable of grasping complex contextual information, harmful biases are likely increasingly intertwined with those models. This paper analyses gender bias in BERT models with two main contributions: First, a novel bias measure is introduced, defining biases as the difference in sentiment valuation of female and male sample versions. Second, we comprehensively analyse BERT's biases on the example of a realistic IMDB movie classifier. By systematically varying elements of the training pipeline, we can conclude regarding their impact on the final model bias. Seven different public BERT models in nine training conditions, i.e. 63 models in total, are compared. Almost all conditions yield significant gender biases. Results indicate that reflected biases stem from public BERT models rather than task-specific data, emphasising the weight of responsible usage.

References in corpus (1)

• Unmasking Contextual Stereotypes: Measuring and Mitigating BERT's Gender Bias

Cited by in corpus (3)

• Bias Against 93 Stigmatized Groups in Masked Language Models and Downstream Sentiment Classification Tasks
• Investigating the Capabilities and Limitations of Machine Learning for Identifying Bias in English Language Data with Information and Heritage Professionals
• GECOBench: A Gender-Controlled Text Dataset and Benchmark for Quantifying Biases in Explanations