Developments in Propaganda and Bias Detection

The field of propaganda and bias detection is moving towards a more nuanced understanding of the role of language and media in shaping public opinion. Researchers are developing new methodologies to analyze the language used by journalists and news outlets, and to detect propaganda and biased content in social media and news articles. The use of large language models is becoming increasingly prevalent in this field, with studies showing their potential to improve the detection of propaganda and biased content. However, the environmental and financial costs of using these models are also being considered. Noteworthy papers in this area include: Analysis of Propaganda in Tweets From Politically Biased Sources, which introduces a new dataset and methodology for analyzing the role of journalists in propagating bias. Toxicity in State Sponsored Information Operations, which presents a comprehensive analysis of toxic language deployment in state-sponsored information operations. Journalism-Guided Agentic In-Context Learning for News Stance Detection, which introduces a new framework for detecting the stance of news articles and promoting viewpoint diversity in news recommendations.

Sources

Analysis of Propaganda in Tweets From Politically Biased Sources

Te Ahorr\'e Un Click: A Revised Definition of Clickbait and Detection in Spanish News

Toxicity in State Sponsored Information Operations

Journalism-Guided Agentic In-Context Learning for News Stance Detection

Fine-Grained Chinese Hate Speech Understanding: Span-Level Resources, Coded Term Lexicon, and Enhanced Detection Frameworks

Social and Political Framing in Search Engine Results

Built with on top of