A Graph-Enhanced Defense Framework for Explainable Fake News Detection with LLM

TL;DR

This paper introduces G-Defense, a graph-based framework utilizing LLMs and retrieval-augmented generation to improve explainable fake news detection with fine-grained, comprehensive explanations.

Contribution

The paper proposes a novel graph-enhanced framework that leverages unverified reports and LLMs to provide detailed explanations and improve detection accuracy.

Findings

G-Defense achieves state-of-the-art accuracy in fake news detection.

It produces high-quality, comprehensive explanations for claims.

The framework effectively models claim dependencies using a graph structure.

Abstract

Explainable fake news detection aims to assess the veracity of news claims while providing human-friendly explanations. Existing methods incorporating investigative journalism are often inefficient and struggle with breaking news. Recent advances in large language models (LLMs) enable leveraging externally retrieved reports as evidence for detection and explanation generation, but unverified reports may introduce inaccuracies. Moreover, effective explainable fake news detection should provide a comprehensible explanation for all aspects of a claim to assist the public in verifying its accuracy. To address these challenges, we propose a graph-enhanced defense framework (G-Defense) that provides fine-grained explanations based solely on unverified reports. Specifically, we construct a claim-centered graph by decomposing the news claim into several sub-claims and modeling their dependency relationships. For each sub-claim, we use the retrieval-augmented generation (RAG) technique to retrieve salient evidence and generate competing explanations. We then introduce a defense-like inference module based on the graph to assess the overall veracity. Finally, we prompt an LLM to generate an intuitive explanation graph. Experimental results demonstrate that G-Defense achieves state-of-the-art performance in both veracity detection and the quality of its explanations.