To address bias in language models, researchers are leveraging established social psychology research on stereotyping. This interdisciplinary approach uses frameworks like the Stereotype Content Model (SCM) to understand how stereotypes about social groups are formed and perpetuated. The SCM posits that stereotypes are based on two dimensions: warmth (intent to harm) and competence (ability to harm). This framework has been applied in NLP for various tasks, including stereotype identification, bias mitigation, and hate speech detection. While the SCM has been extensively studied in English language models and Western cultural contexts, its applicability as a cross-cultural measure of stereotypes remains an open research question. This paper explores the cross-cultural validity of the SCM by developing a Korean Stereotype Content Model (KoSCM). We create a Korean warmth-competence lexicon through machine translation of existing English lexicons, validated by an expert translator, and utilize this lexicon to develop a labeled training dataset of Korean sentences. This work presents the first extension of SCM lexicons to a non-English language (Korean), aiming to broaden understanding of stereotypes and cultural dynamics.

Natural Language Processing (NLP) models tend to inherit and amplify stereotypical biases present in their training data, leading to harmful societal consequences. Current efforts to rectify these biases typically revolve around making models oblivious to bias, which is at odds with the idea that humans require increased awareness to tackle these biases better. This prompts a fundamental research question: are bias-oblivious models the only viable solution to combat stereotypical biases? This paper answers this question by proposing the Agency-BeLiefs Embedding (ABLE) model, a novel approach that actively encodes stereotypical biases into the embedding space. ABLE draws upon social psychological theory to acquire and represent stereotypical biases in the form of agency and belief scores rather than directly representing stereotyped groups. Our experimental results showcase ABLE’s effectiveness in learning agency and belief stereotypes while preserving the language model’s proficiency. Furthermore, we underscore the practical significance of incorporating stereotypes within the ABLE model by demonstrating its utility in various downstream tasks. Our approach exemplifies the potential benefits of addressing bias through awareness, as opposed to the prevailing approach of mitigating bias through obliviousness.

Framing is a political strategy in which journalists and politicians emphasize certain aspects of a societal issue in order to influence and sway public opinion. Frameworks for detecting framing in news articles or social media posts are critical in understanding the spread of biased information in our society. In this paper, we propose CLoSE, a multi-task BERT-based model which uses contrastive learning to embed indicators of frames from news articles in order to predict political bias. We evaluate the performance of our proposed model on subframes and political bias classification tasks. We also demonstrate the model’s classification accuracy on zero-shot and few-shot learning tasks, providing a promising avenue for framing detection in unlabeled data.