Discriminatory Language

Discriminatory language research focuses on identifying and mitigating bias in text and other media, aiming to create fairer and more inclusive online and offline environments. Current research utilizes various approaches, including natural language processing models like BERT and LLMs such as Llama 2, often incorporating multimodal data (text and images) and leveraging techniques like fine-tuning and bias detection algorithms to identify subtle forms of prejudice, including microaggressions and dialect prejudice. This work is crucial for addressing societal inequalities and improving the fairness and safety of AI systems, with implications for content moderation, hiring practices, and legal enforcement of anti-discrimination laws.