The research paper introduces FACT-GPT, a system that uses Large Language Models (LLMs) to automate the fact-checking process by matching claims on social media to previously debunked claims. The authors argue that this approach mirrors human judgment in accuracy and can help fact-checkers by reducing redundant verification, assisting in content moderation, and supporting the analysis of misinformation.
Publication date: 8 Feb 2024
Project Page: https://arxiv.org/abs/2402.05904v1
Paper: https://arxiv.org/pdf/2402.05904
