Unsupervised Cross-Domain Rumor Detection with Contrastive Learning and Cross-Attention

TL;DR

This paper introduces a novel unsupervised cross-domain rumor detection model that leverages contrastive learning and cross-attention to improve detection accuracy across different domains without requiring labeled target data.

Contribution

It proposes an end-to-end contrastive learning framework with cross-attention for domain-invariant rumor detection, utilizing pseudo labels and prototype alignment.

Findings

Achieves state-of-the-art performance on four cross-domain rumor datasets.

Effectively aligns features across domains without target labels.

Improves detection accuracy in cross-domain scenarios.

Abstract

Massive rumors usually appear along with breaking news or trending topics, seriously hindering the truth. Existing rumor detection methods are mostly focused on the same domain, and thus have poor performance in cross-domain scenarios due to domain shift. In this work, we propose an end-to-end instance-wise and prototype-wise contrastive learning model with a cross-attention mechanism for cross-domain rumor detection. The model not only performs cross-domain feature alignment but also enforces target samples to align with the corresponding prototypes of a given source domain. Since target labels in a target domain are unavailable, we use a clustering-based approach with carefully initialized centers by a batch of source domain samples to produce pseudo labels. Moreover, we use a cross-attention mechanism on a pair of source data and target data with the same labels to learn domain-invariant representations. Because the samples in a domain pair tend to express similar semantic patterns, especially on the people's attitudes (e.g., supporting or denying) towards the same category of rumors, the discrepancy between a pair of the source domain and target domain will be decreased. We conduct experiments on four groups of cross-domain datasets and show that our proposed model achieves state-of-the-art performance.