Exposing Pink Slime Journalism: Linguistic Signatures and Robust Detection Against LLM-Generated Threats
This addresses the threat of low-quality, auto-generated news for 28 million Americans relying on local news, with incremental improvements in detection robustness.
The study tackled the problem of detecting Pink Slime Journalism, a type of auto-generated deceptive local news, by analyzing its linguistic patterns and proposing detection strategies, revealing that LLM-based attacks reduce existing detection performance by up to 40% in F1-score, while their robust framework improved detection by up to 27%.
The local news landscape, a vital source of reliable information for 28 million Americans, faces a growing threat from Pink Slime Journalism, a low-quality, auto-generated articles that mimic legitimate local reporting. Detecting these deceptive articles requires a fine-grained analysis of their linguistic, stylistic, and lexical characteristics. In this work, we conduct a comprehensive study to uncover the distinguishing patterns of Pink Slime content and propose detection strategies based on these insights. Beyond traditional generation methods, we highlight a new adversarial vector: modifications through large language models (LLMs). Our findings reveal that even consumer-accessible LLMs can significantly undermine existing detection systems, reducing their performance by up to 40% in F1-score. To counter this threat, we introduce a robust learning framework specifically designed to resist LLM-based adversarial attacks and adapt to the evolving landscape of automated pink slime journalism, and showed and improvement by up to 27%.