Hierarchical Multi-head Attentive Network for Evidence-aware Fake News Detection

TL;DR

This paper introduces a hierarchical multi-head attentive network that effectively combines word-level and evidence-level attention to improve fake news detection accuracy using external evidence, outperforming existing methods.

Contribution

The novel hierarchical multi-head attentive network jointly models word and evidence attention, enhancing interpretability and detection performance in fake news verification.

Findings

Outperforms seven state-of-the-art baselines

Achieves 6% to 18% improvements over baselines

Effective in both word-level and evidence-level explanations

Abstract

The widespread of fake news and misinformation in various domains ranging from politics, economics to public health has posed an urgent need to automatically fact-check information. A recent trend in fake news detection is to utilize evidence from external sources. However, existing evidence-aware fake news detection methods focused on either only word-level attention or evidence-level attention, which may result in suboptimal performance. In this paper, we propose a Hierarchical Multi-head Attentive Network to fact-check textual claims. Our model jointly combines multi-head word-level attention and multi-head document-level attention, which aid explanation in both word-level and evidence-level. Experiments on two real-word datasets show that our model outperforms seven state-of-the-art baselines. Improvements over baselines are from 6\% to 18\%. Our source code and datasets are released at \texttt{\url{https://github.com/nguyenvo09/EACL2021}}.