The Perils & Promises of Fact-checking with Large Language Models

TL;DR

This paper evaluates the capabilities and limitations of Large Language Models like GPT-4 in automated fact-checking, emphasizing the importance of context, source citation, and cautious application due to inconsistent accuracy.

Contribution

It introduces a framework for LLM-based fact-checking that includes reasoning and source citation, and compares GPT-4's performance to GPT-3 across various conditions.

Findings

GPT-4 outperforms GPT-3 in fact-checking accuracy

Contextual information significantly improves LLM performance

Accuracy varies with query language and claim veracity

Abstract

Automated fact-checking, using machine learning to verify claims, has grown vital as misinformation spreads beyond human fact-checking capacity. Large Language Models (LLMs) like GPT-4 are increasingly trusted to write academic papers, lawsuits, and news articles and to verify information, emphasizing their role in discerning truth from falsehood and the importance of being able to verify their outputs. Understanding the capacities and limitations of LLMs in fact-checking tasks is therefore essential for ensuring the health of our information ecosystem. Here, we evaluate the use of LLM agents in fact-checking by having them phrase queries, retrieve contextual data, and make decisions. Importantly, in our framework, agents explain their reasoning and cite the relevant sources from the retrieved context. Our results show the enhanced prowess of LLMs when equipped with contextual information. GPT-4 outperforms GPT-3, but accuracy varies based on query language and claim veracity. While LLMs show promise in fact-checking, caution is essential due to inconsistent accuracy. Our investigation calls for further research, fostering a deeper comprehension of when agents succeed and when they fail.