Uncovering Implicit Gender Bias in Narratives through Commonsense Inference

Abstract

Pre-trained language models learn socially harmful biases from their training corpora, and may repeat these biases when used for generation. We study gender biases associated with the protagonist in model-generated stories. Such biases may be expressed either explicitly ("women can't park") or implicitly (e.g. an unsolicited male character guides her into a parking space). We focus on implicit biases, and use a commonsense reasoning engine to uncover them. Specifically, we infer and analyze the protagonist's motivations, attributes, mental states, and implications on others. Our findings regarding implicit biases are in line with prior work that studied explicit biases, for example showing that female characters' portrayal is centered around appearance, while male figures' focus on intellect.

arXiv commentAccepted at Findings of EMNLP 2021

References in corpus (4)

• Efficient Estimation of Word Representations in Vector Space
• Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings
• Empath: Understanding Topic Signals in Large-Scale Text
• Shirtless and Dangerous: Quantifying Linguistic Signals of Gender Bias in an Online Fiction Writing Community