Adapting Fake News Detection to the Era of Large Language Models

TL;DR

This paper investigates how fake news detection models can be adapted to the era of large language models, revealing that training on human-written news helps detect machine-generated fake news, and proposing strategies for robustness.

Contribution

It provides a comprehensive evaluation of fake news detectors across different scenarios involving human and machine-generated news, and offers practical adaptation strategies.

Findings

Detectors trained on human-written news perform well on machine-generated fake news.

Detectors trained solely on machine-generated news are less effective.

Training with a lower ratio of machine-generated news improves detection robustness.

Abstract

In the age of large language models (LLMs) and the widespread adoption of AI-driven content creation, the landscape of information dissemination has witnessed a paradigm shift. With the proliferation of both human-written and machine-generated real and fake news, robustly and effectively discerning the veracity of news articles has become an intricate challenge. While substantial research has been dedicated to fake news detection, this either assumes that all news articles are human-written or abruptly assumes that all machine-generated news are fake. Thus, a significant gap exists in understanding the interplay between machine-(paraphrased) real news, machine-generated fake news, human-written fake news, and human-written real news. In this paper, we study this gap by conducting a comprehensive evaluation of fake news detectors trained in various scenarios. Our primary objectives revolve around the following pivotal question: How to adapt fake news detectors to the era of LLMs? Our experiments reveal an interesting pattern that detectors trained exclusively on human-written articles can indeed perform well at detecting machine-generated fake news, but not vice versa. Moreover, due to the bias of detectors against machine-generated texts \cite{su2023fake}, they should be trained on datasets with a lower machine-generated news ratio than the test set. Building on our findings, we provide a practical strategy for the development of robust fake news detectors.