MEMEWEAVER: Inter-Meme Graph Reasoning for Sexism and Misogyny Detection

TL;DR

MemeWeaver is a novel multimodal framework that uses inter-meme graph reasoning to improve detection of online sexism and misogyny, capturing social dynamics and outperforming existing methods.

Contribution

It introduces an end-to-end trainable inter-meme graph reasoning mechanism and systematically evaluates multimodal fusion strategies for better hate speech detection.

Findings

Outperforms state-of-the-art baselines on MAMI and EXIST datasets.

Achieves faster training convergence.

Learned graph structures reveal meaningful social patterns.

Abstract

Women are twice as likely as men to face online harassment due to their gender. Despite recent advances in multimodal content moderation, most approaches still overlook the social dynamics behind this phenomenon, where perpetrators reinforce prejudices and group identity within like-minded communities. Graph-based methods offer a promising way to capture such interactions, yet existing solutions remain limited by heuristic graph construction, shallow modality fusion, and instance-level reasoning. In this work, we present MemeWeaver, an end-to-end trainable multimodal framework for detecting sexism and misogyny through a novel inter-meme graph reasoning mechanism. We systematically evaluate multiple visual--textual fusion strategies and show that our approach consistently outperforms state-of-the-art baselines on the MAMI and EXIST benchmarks, while achieving faster training convergence. Further analyses reveal that the learned graph structure captures semantically meaningful patterns, offering valuable insights into the relational nature of online hate.