Two Birds with One Stone: Improving Rumor Detection by Addressing the Unfairness Issue

TL;DR

This paper introduces a two-step framework for rumor detection that identifies confounding sensitive attributes and learns fair, invariant representations, significantly enhancing detection accuracy and fairness without needing sensitive attribute labels.

Contribution

The novel approach simultaneously improves rumor detection performance and fairness by identifying confounding attributes and applying invariant learning without sensitive attribute annotations.

Findings

Significant improvement in rumor detection accuracy.

Enhanced fairness across different groups.

Easy integration with existing detectors.

Abstract

The degraded performance and group unfairness caused by confounding sensitive attributes in rumor detection remains relatively unexplored. To address this, we propose a two-step framework. Initially, it identifies confounding sensitive attributes that limit rumor detection performance and cause unfairness across groups. Subsequently, we aim to learn equally informative representations through invariant learning. Our method considers diverse sets of groups without sensitive attribute annotations. Experiments show our method easily integrates with existing rumor detectors, significantly improving both their detection performance and fairness.