Evaluating the Factuality of Large Language Models using Large-Scale Knowledge Graphs

TL;DR

This paper introduces GraphEval, a scalable method for assessing the factual accuracy of large language models using a vast knowledge graph, reducing evaluation costs and improving alignment with true correctness.

Contribution

We propose GraphEval, a novel evaluation framework that leverages a large knowledge graph and a judge model to efficiently and accurately assess LLM factuality.

Findings

Judge model's assessments closely match actual correctness.

Evaluation process is significantly more cost-effective.

Insights into LLM performance metrics and potential improvements.

Abstract

The advent of Large Language Models (LLMs) has significantly transformed the AI landscape, enhancing machine learning and AI capabilities. Factuality issue is a critical concern for LLMs, as they may generate factually incorrect responses. In this paper, we propose GraphEval to evaluate an LLM's performance using a substantially large test dataset. Specifically, the test dataset is retrieved from a large knowledge graph with more than 10 million facts without expensive human efforts. Unlike conventional methods that evaluate LLMs based on generated responses, GraphEval streamlines the evaluation process by creating a judge model to estimate the correctness of the answers given by the LLM. Our experiments demonstrate that the judge model's factuality assessment aligns closely with the correctness of the LLM's generated outputs, while also substantially reducing evaluation costs. Besides, our findings offer valuable insights into LLM performance across different metrics and highlight the potential for future improvements in ensuring the factual integrity of LLM outputs. The code is publicly available at https://github.com/xz-liu/GraphEval.