发表机构
Carnegie Mellon University; RMIT University; Yale University（卡内基梅隆大学; 皇家墨尔本理工大学; 耶鲁大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出多智能体框架生成多模态假新闻，并基准测试16个MLLM的检测能力，发现多数模型远低于人类水平，尤其在图像真实性识别上失败，为防御假新闻奠定基础。
AI 中文摘要
生成式人工智能的快速发展引发了对多模态大语言模型（MLLMs）被滥用于社交媒体大规模虚假信息活动的担忧。尽管已有关于文本虚假信息的研究，但一个根本性问题仍未得到解答：多模态大语言模型能否被利用来制造逼真的多模态假新闻，以及它们能否可靠地检测出这些假新闻？我们引入了一个多智能体框架，其中故事智能体、图像智能体和评论智能体协作生成看似合理地反驳真实新闻的虚假社交媒体帖子。我们应用该框架在科学、健康和娱乐领域生成了超过9,000对多模态新闻帖子，并基准测试了16个开源和闭源多模态大语言模型进行自动检测。我们发现，大多数模型在准确率上远低于人类水平，并且在识别图像真实性方面严重失败。我们的研究为开发针对社交媒体假新闻的稳健防御措施奠定了基础。代码和数据可在https://github.com/xiuzhenzhang/Multimodal获取。
英文摘要
The rapid advancement of generative AI raises concerns about the misuse of Multimodal LLMs (MLLMs) for large-scale disinformation campaigns on social media. Despite existing research on textual disinformation, a fundamental question remains unanswered: can MLLMs be exploited to fabricate realistic multimodal fake news, and can they reliably detect it? We introduce a multi-agent framework in which a story agent, an image agent, and a critic agent collaborate to produce fake social media posts that plausibly counter true news. We apply the framework to generate over 9,000 paired multimodal news posts across science, health, and entertainment domains, and benchmark 16 open- and closed-source MLLMs for automated detection. We find that most models fall substantially short of human-level accuracy and fail critically on identifying image authenticity. Our research provides a foundation for developing robust defenses against social media fake news. Code and data are available at https: //github.com/xiuzhenzhang/Multimodal.
Comments15 pages, 6 figures. Accepted for publication in the Findings of EMNLP 2026
