Agentic Mixed-Source Multi-Modal Misinformation Detection with Adaptive Test-Time Scaling

TL;DR

This paper introduces AgentM3D, a multi-agent framework that enhances zero-shot multi-modal misinformation detection by employing adaptive reasoning strategies and dynamic path exploration, significantly improving detection accuracy.

Contribution

The work proposes a novel multi-agent system with adaptive test-time scaling and reasoning path exploration to improve zero-shot multi-modal misinformation detection.

Findings

Achieves state-of-the-art zero-shot detection performance on M3D benchmarks.

Outperforms existing VLM-based and agentic baselines.

Demonstrates effectiveness of adaptive reasoning in complex misinformation tasks.

Abstract

Vision-language models (VLMs) have been proven effective for detecting multi-modal misinformation on social platforms, especially in zero-shot settings with unavailable or delayed annotations. However, a single VLM's capacity falls short in the more complex mixed-source multi-modal misinformation detection (M3D) task. Taking captioned images as an example, in M3D, false information can originate from untruthful texts, forged images, or mismatches between the two modalities. Although recent agentic systems can handle zero-shot M3D by connecting modality-specific VLM agents, their effectiveness is still bottlenecked by their architecture. In existing agentic M3D solutions, for any input sample, each agent performs only one forward reasoning pass, making decisions prone to model randomness and reasoning errors in challenging cases. Moreover, the lack of exploration over alternative reasoning paths prevents modern VLMs from fully utilizing their reasoning capacity. In this work, we present AgentM3D, a multi-agent framework for zero-shot M3D. To amplify the reasoning capability of VLMs, we introduce an adaptive test-time scaling paradigm in which each modality-specific VLM agent applies a Best-of-N mechanism, coupled with a critic agent for task-aligned scoring. The agents are organized in a cascading, modality-specific decision chain to reduce unnecessary computation and limit error propagation. To ensure scalability, a planning agent dynamically determines the maximum number of reasoning paths based on sample difficulty, and an adaptive stopping mechanism prevents excessive reasoning within each agent. Extensive experiments on two M3D benchmarks demonstrate that AgentM3D achieves state-of-the-art zero-shot detection performance compared with various VLM-based and agentic baselines.