Religious Bias Landscape in Language and Text-to-Image Models: Analysis, Detection, and Debiasing Strategies
It addresses the problem of inherent religious biases in AI models for developers and users, highlighting an urgent need for fairer systems, though it is incremental as it builds on existing bias analysis work.
This study systematically investigated religious bias in language and text-to-image models, revealing concerning stereotypes and biases disproportionately affecting certain religions through approximately 400 prompts, and found that targeted debiasing techniques like corrective prompts can mitigate these biases.
Note: This paper includes examples of potentially offensive content related to religious bias, presented solely for academic purposes. The widespread adoption of language models highlights the need for critical examinations of their inherent biases, particularly concerning religion. This study systematically investigates religious bias in both language models and text-to-image generation models, analyzing both open-source and closed-source systems. We construct approximately 400 unique, naturally occurring prompts to probe language models for religious bias across diverse tasks, including mask filling, prompt completion, and image generation. Our experiments reveal concerning instances of underlying stereotypes and biases associated disproportionately with certain religions. Additionally, we explore cross-domain biases, examining how religious bias intersects with demographic factors such as gender, age, and nationality. This study further evaluates the effectiveness of targeted debiasing techniques by employing corrective prompts designed to mitigate the identified biases. Our findings demonstrate that language models continue to exhibit significant biases in both text and image generation tasks, emphasizing the urgent need to develop fairer language models to achieve global acceptability.