MV-Debate: Multi-view Agent Debate with Dynamic Reflection Gating for Multimodal Harmful Content Detection in Social Media

TL;DR

This paper introduces MV-Debate, a multi-view agent debate framework with dynamic reflection gating, to improve multimodal harmful content detection on social media by leveraging diverse interpretive perspectives and iterative refinement.

Contribution

The paper presents a novel multi-agent debate framework with dynamic reflection gating that enhances multimodal harmful content detection beyond existing methods.

Findings

Outperforms existing single-model and multi-agent debate baselines.

Effectively analyzes multimodal content with diverse interpretive perspectives.

Demonstrates significant improvements on benchmark datasets.

Abstract

Social media has evolved into a complex multimodal environment where text, images, and other signals interact to shape nuanced meanings, often concealing harmful intent. Identifying such intent, whether sarcasm, hate speech, or misinformation, remains challenging due to cross-modal contradictions, rapid cultural shifts, and subtle pragmatic cues. To address these challenges, we propose MV-Debate, a multi-view agent debate framework with dynamic reflection gating for unified multimodal harmful content detection. MV-Debate assembles four complementary debate agents, a surface analyst, a deep reasoner, a modality contrast, and a social contextualist, to analyze content from diverse interpretive perspectives. Through iterative debate and reflection, the agents refine responses under a reflection-gain criterion, ensuring both accuracy and efficiency. Experiments on three benchmark datasets demonstrate that MV-Debate significantly outperforms strong single-model and existing multi-agent debate baselines. This work highlights the promise of multi-agent debate in advancing reliable social intent detection in safety-critical online contexts.