KG-FPQ: Evaluating Factuality Hallucination in LLMs with Knowledge Graph-based False Premise Questions

TL;DR

This paper introduces KG-FPQ, a large-scale, automated benchmark for evaluating how susceptible large language models are to factuality hallucination caused by false premise questions, leveraging knowledge graphs and GPTs.

Contribution

It presents a novel scalable pipeline for generating false premise questions from knowledge graphs and creates the first extensive benchmark dataset for evaluating LLM vulnerability to factual hallucination.

Findings

LLMs show varying susceptibility to false premise questions.

The benchmark reveals specific weaknesses in current LLMs.

The dataset enables systematic evaluation of factuality hallucination.

Abstract

Recent studies have demonstrated that large language models (LLMs) are susceptible to being misled by false premise questions (FPQs), leading to errors in factual knowledge, know as factuality hallucination. Existing benchmarks that assess this vulnerability primarily rely on manual construction, resulting in limited scale and lack of scalability. In this work, we introduce an automated, scalable pipeline to create FPQs based on knowledge graphs (KGs). The first step is modifying true triplets extracted from KGs to create false premises. Subsequently, utilizing the state-of-the-art capabilities of GPTs, we generate semantically rich FPQs. Based on the proposed method, we present a comprehensive benchmark, the Knowledge Graph-based False Premise Questions (KG-FPQ), which contains approximately 178k FPQs across three knowledge domains, at six levels of confusability, and in two task formats. Using KG-FPQ, we conduct extensive evaluations on several representative LLMs and provide valuable insights. The KG-FPQ dataset and code are available at~https://github.com/yanxuzhu/KG-FPQ.