ArMeme: Propagandistic Content in Arabic Memes

TL;DR

This paper introduces ArMeme, a new Arabic memes dataset with annotations for propagandistic content, aiming to facilitate detection of misleading memes in a low-resource language context.

Contribution

It presents the first Arabic multimodal memes dataset with manual annotations for propagandistic content, enabling research in Arabic meme misinformation detection.

Findings

Annotated 6,000 Arabic memes for propagandistic content.

Provided a comprehensive analysis of Arabic memes for detection.

Made the dataset publicly available for research.

Abstract

With the rise of digital communication, memes have become a significant medium for cultural and political expression that is often used to mislead audiences. Identification of such misleading and persuasive multimodal content has become more important among various stakeholders, including social media platforms, policymakers, and the broader society as they often cause harm to individuals, organizations, and/or society. While there has been effort to develop AI-based automatic systems for resource-rich languages (e.g., English), it is relatively little to none for medium to low resource languages. In this study, we focused on developing an Arabic memes dataset with manual annotations of propagandistic content. We annotated ~6K Arabic memes collected from various social media platforms, which is a first resource for Arabic multimodal research. We provide a comprehensive analysis aiming to develop computational tools for their detection. We will make them publicly available for the community.