MemeIntel: Explainable Detection of Propagandistic and Hateful Memes

ACL ARR 2025 February Submission4835 Authors

Abstract: The proliferation of multimodal content on social media presents significant challenges in understanding and moderating complex, context-dependent issues such as misinformation, hate speech, and propaganda. While efforts have been made to develop resources and propose new methods for automatic detection, limited attention has been given to label detection and the generation of explanation-based rationales for predicted labels. To address this challenge, we introduce MemeXplain, an explanation-enhanced dataset for propaganda memes in Arabic and hateful memes in English, making it the first large-scale resource for these tasks. To tackle these challenges, we propose a multi-stage optimization approach and train Vision-Language Models (VLMs). Our results show that this approach significantly improves performance over the base model for both label detection and explanation generation, surpassing the current state-of-the-art with an absolute improvement of ~3% on ArMeme and ~7% on Hateful Memes. For reproducibility and future research, we aim to make the MemeXplain dataset and experimental resources publicly available (anonymous.com).

Paper Type: Long

Research Area: Resources and Evaluation

Research Area Keywords: misinformation detection and analysis, hateful meme detection, corpus creation, datasets for low resource languages

Contribution Types: Approaches to low-resource settings, Data resources, Data analysis

Languages Studied: Arabic, English

Submission Number: 4835