Fake News Detection
Overview
Fake news detection studies how to identify false or misleading news, claims, and media, helping people and platforms make better-informed decisions. Methods range from text classifiers and social-network or propagation models to multimodal systems that compare text, images, and video, often using retrieved evidence, knowledge graphs, or large language models for verification. Researchers also use human-in-the-loop labeling, explanations, and benchmarks spanning languages, domains, and forms of manipulation. A central challenge is that strong results on familiar datasets may not transfer to new events, languages, or adversarially rewritten content, while evidence quality, false alarms, and uncertainty remain difficult to manage.
Latest papers
98 papers12 papers in the last four weeks, up 50% on the four weeks before. 0.1% of all new papers.