Leveraging Knowledge Graphs and LLMs for Structured Generation of Misinformation

TL;DR

This paper introduces a method combining knowledge graphs and large language models to generate and analyze structured misinformation, revealing limitations in current detection techniques and emphasizing the need for improved strategies.

Contribution

It presents a novel approach that uses knowledge graphs to systematically generate realistic misinformation and evaluates LLMs' ability to detect such fabricated data.

Findings

Generated misinformation is difficult for humans to detect.

Current LLM detection methods have significant limitations.

Structured semantic relationships can guide more realistic misinformation generation.

Abstract

The rapid spread of misinformation, further amplified by recent advances in generative AI, poses significant threats to society, impacting public opinion, democratic stability, and national security. Understanding and proactively assessing these threats requires exploring methodologies that enable structured and scalable misinformation generation. In this paper, we propose a novel approach that leverages knowledge graphs (KGs) as structured semantic resources to systematically generate fake triplets. By analyzing the structural properties of KGs, such as the distance between entities and their predicates, we identify plausibly false relationships. These triplets are then used to guide large language models (LLMs) in generating misinformation statements with varying degrees of credibility. By utilizing structured semantic relationships, our deterministic approach produces misinformation inherently challenging for humans to detect, drawing exclusively upon publicly available KGs (e.g., WikiGraphs). Additionally, we investigate the effectiveness of LLMs in distinguishing between genuine and artificially generated misinformation. Our analysis highlights significant limitations in current LLM-based detection methods, underscoring the necessity for enhanced detection strategies and a deeper exploration of inherent biases in generative models.