FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning

TL;DR

FactGuard is an innovative agentic framework that enhances video misinformation detection by iterative reasoning and external evidence verification, outperforming existing models in accuracy and robustness.

Contribution

The paper introduces FactGuard, a novel agentic approach utilizing reinforcement learning and external tool invocation for improved video misinformation detection.

Findings

Achieves state-of-the-art results on FakeSV, FakeTT, and FakeVV datasets.

Demonstrates robustness and generalization across diverse video misinformation scenarios.

Employs a two-stage training strategy combining supervised fine-tuning and reinforcement learning.

Abstract

Multimodal large language models (MLLMs) have substantially advanced video misinformation detection through unified multimodal reasoning, but they often rely on fixed-depth inference and place excessive trust in internally generated assumptions, particularly in scenarios where critical evidence is sparse, fragmented, or requires external verification. To address these limitations, we propose FactGuard, an agentic framework for video misinformation detection that formulates verification as an iterative reasoning process built upon MLLMs. FactGuard explicitly assesses task ambiguity and selectively invokes external tools to acquire critical evidence, enabling progressive refinement of reasoning trajectories. To further strengthen this capability, we introduce a two-stage training strategy that combines domain-specific agentic supervised fine-tuning with decision-aware reinforcement learning to optimize tool usage and calibrate risk-sensitive decision making. Extensive experiments on FakeSV, FakeTT, and FakeVV demonstrate FactGuard's state-of-the-art performance and validate its excellent robustness and generalization capacity.