Medium Bias

Media bias research aims to automatically detect and classify biases in news articles and social media, focusing on identifying various bias types (political, racial, gender, etc.) and their interrelationships across different domains. Current research leverages large language models (LLMs), such as BERT and RoBERTa, often employing multi-task learning and domain adaptation techniques to improve accuracy and generalizability across diverse datasets. This work is crucial for promoting more informed media consumption and fostering a more equitable public discourse by providing tools to analyze and understand the pervasive influence of bias in information dissemination.