Fake News Detection After LLM Laundering: Measurement and Explanation
The research tested how well fake news detectors identify misinformation when it is rewritten by Large Language Models (LLMs). Findings show detectors struggle more with LLM-paraphrased text than human text, suggesting LLM laundering can evade detection. A key reason for failure is a shift in sentiment or the introduction of ambiguity during paraphrasing.