LLM-GAN: Construct Generative Adversarial Network Through Large Language Models For Explainable Fake News Detection

2024-09-03

arXiv.org

AI Summary

Existing fake news detection methods underutilize large language models (LLMs), suffering from low accuracy, poor interpretability, high misclassification rates on highly realistic fake news, and heavy reliance on manual verification—resulting in low efficiency. To address these limitations, we propose an LLM-intrinsic adversarial paradigm: leveraging prompt engineering to endow a single LLM with dual roles—as both a generator and an interpretable detector—thereby establishing a closed-loop framework of generation–detection–explanation. Our approach integrates generative adversarial training, interpretability alignment mechanisms, and a cloud-native service architecture. Evaluated on multiple benchmark datasets, our method achieves an 8.2% improvement in prediction accuracy and attains an explanation faithfulness score of 0.87 against human evaluations. The solution has been deployed as a real-time, interpretable fake news detection cloud service.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Application Domains: Misinformation & Fake News

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Explainable fake news detection predicts the authenticity of news items with annotated explanations. Today, Large Language Models (LLMs) are known for their powerful natural language understanding and explanation generation abilities. However, presenting LLMs for explainable fake news detection remains two main challenges. Firstly, fake news appears reasonable and could easily mislead LLMs, leaving them unable to understand the complex news-faking process. Secondly, utilizing LLMs for this task would generate both correct and incorrect explanations, which necessitates abundant labor in the loop. In this paper, we propose LLM-GAN, a novel framework that utilizes prompting mechanisms to enable an LLM to become Generator and Detector and for realistic fake news generation and detection. Our results demonstrate LLM-GAN's effectiveness in both prediction performance and explanation quality. We further showcase the integration of LLM-GAN to a cloud-native AI platform to provide better fake news detection service in the cloud.

Problem

• fake news detection
• ultra-large language models
• interpretability

Innovation

• LLM-GAN
• Bidirectional Capability
• Cloud Integration