An Experimental Comparison of the Most Popular Approaches to Fake News Detection

TL;DR

This paper critically compares 12 fake news detection methods across multiple datasets, highlighting the strengths and limitations of traditional, deep learning, transformer, and LLM approaches in real-world scenarios.

Contribution

It provides a comprehensive evaluation of diverse detection approaches on varied datasets, emphasizing domain shift challenges and the potential of LLMs for zero- and few-shot detection.

Findings

Fine-tuned models excel in-domain but struggle cross-domain.

Cross-domain architectures reduce generalization gap but require lots of data.

LLMs show promise with zero- and few-shot learning capabilities.

Abstract

In recent years, fake news detection has received increasing attention in public debate and scientific research. Despite advances in detection techniques, the production and spread of false information have become more sophisticated, driven by Large Language Models (LLMs) and the amplification power of social media. We present a critical assessment of 12 representative fake news detection approaches, spanning traditional machine learning, deep learning, transformers, and specialized cross-domain architectures. We evaluate these methods on 10 publicly available datasets differing in genre, source, topic, and labeling rationale. We address text-only English fake news detection as a binary classification task by harmonizing labels into "Real" and "Fake" to ensure a consistent evaluation protocol. We acknowledge that label semantics vary across datasets and that harmonization inevitably removes such semantic nuances. Each dataset is treated as a distinct domain. We conduct in-domain, multi-domain and cross-domain experiments to simulate real-world scenarios involving domain shift and out-of-distribution data. Fine-tuned models perform well in-domain but struggle to generalize. Cross-domain architectures can reduce this gap but are data-hungry, while LLMs offer a promising alternative through zero- and few-shot learning. Given inherent dataset confounds and possible pre-training exposure, results should be interpreted as robustness evaluations within this English, text-only protocol.