Generating Realistic Adversarial User Comment Attacks to Evaluate the Robustness of Fake News Detectors

Abstract: The rise of social media has amplified the spread of fake news, threatening societal stability. To address this, machine learning-based fake news detection systems analyze news content, including text, images, and user interactions. However, these systems are vulnerable to attacks that manipulate data, undermining their effectiveness…we focus on generating more realistic comments for specific news articles. We propose three novel attack strategies: two fine-tune a small language model using reinforcement learning algorithms transformer reinforcement learning and reinforced self-training, while the third uses a large language model with self-reflection. We conduct experiments showing that our methods outperform the state-of-the-art MALCOM procedure and that our proposed attack strategies produce more realistic comments, which are consequently more difficult to detect.

Evaluating the robustness of fake news detectors to adversarial attacks with real user comments

Abstract: The widespread use of social media has led to an increase in false and misleading information presented as legitimate news, also known as fake news. This poses a significant threat to societal stability and has led to the development of fake news detectors that use machine learning to flag suspicious information. These automated systems analyze various aspects of news content, including text, images, and user interactions…

Understanding Teenagers’ Real and Fake News Sharing on Social Media

Abstract: In this paper, we present results from our research work focused on understanding teenagers’ real and fake news sharing on social media. Using existing real/fake news samples and best practices in qualitative, inductive data analysis, we identify factors that explain teenagers’ sharing and not sharing behavior of real and fake news. Our findings suggest that influencing teenagers’ decisions to share or not share news on social media requires changes in the knowledge of what is found in social media and enhancement of knowledge of the algorithmic effects of sharing or not sharing.