![[2602.22963] FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.22963] FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning
Summary
FactGuard introduces an innovative framework for detecting video misinformation using reinforcement learning, enhancing the capabilities of multimodal large language models (MLLMs).
Why It Matters
As misinformation proliferates, especially in video content, effective detection methods are crucial. FactGuard's approach addresses limitations in existing MLLMs by incorporating iterative reasoning and external verification, making it a significant advancement in AI-driven misinformation detection.
Key Takeaways
- FactGuard employs iterative reasoning to improve video misinformation detection.
- The framework utilizes reinforcement learning to optimize decision-making processes.
- Extensive experiments show FactGuard's superior robustness and generalization capabilities.
Computer Science > Artificial Intelligence arXiv:2602.22963 (cs) [Submitted on 26 Feb 2026] Title:FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning Authors:Zehao Li, Hongwei Yu, Hao Jiang, Qiang Sheng, Yilong Xu, Baolong Bi, Yang Li, Zhenlong Yuan, Yujun Cai, Zhaoqi Wang View a PDF of the paper titled FactGuard: Agentic Video Misinformation Detection via Reinforcement Learning, by Zehao Li and 9 other authors View PDF HTML (experimental) 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 Fak...