MultiPress: A Multi-Agent Framework for Interpretable Multimodal News Classification

TL;DR

MultiPress is a multi-agent framework that improves multimodal news classification by integrating perception, reasoning, and fusion agents, leveraging retrieval-augmented reasoning for better accuracy and interpretability.

Contribution

It introduces a novel three-stage multi-agent approach with retrieval-augmented reasoning, advancing multimodal news classification beyond existing methods.

Findings

Significant accuracy improvements over baseline models

Effective modular collaboration among perception, reasoning, and fusion agents

Enhanced interpretability of news classification results

Abstract

With the growing prevalence of multimodal news content, effective news topic classification demands models capable of jointly understanding and reasoning over heterogeneous data such as text and images. Existing methods often process modalities independently or employ simplistic fusion strategies, limiting their ability to capture complex cross-modal interactions and leverage external knowledge. To overcome these limitations, we propose MultiPress, a novel three-stage multi-agent framework for multimodal news classification. MultiPress integrates specialized agents for multimodal perception, retrieval-augmented reasoning, and gated fusion scoring, followed by a reward-driven iterative optimization mechanism. We validate MultiPress on a newly constructed large-scale multimodal news dataset, demonstrating significant improvements over strong baselines and highlighting the effectiveness of modular multi-agent collaboration and retrieval-augmented reasoning in enhancing classification accuracy and interpretability.