发表机构
Tsinghua University; Peking University; Renmin University of China; Microsoft（清华大学; 北京大学; 中国人民大学; 微软公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对AI生成视频检测泛化性不足的问题，首次将元检测引入该领域，提出结合可验证时间证据的VidForensics-M1模型及相关机制，实现了鲁棒可泛化的检测。
AI 中文摘要
近期视频生成模型的进展显著提升了合成视频的真实感，模糊了生成内容与真实内容的边界，引发了对虚假信息传播的担忧。现有基于多模态大语言模型（MLLM）的检测器主要依赖监督微调或标签级强化学习，其中粗粒度的监督限制了其对未见过场景和新兴视频生成器的泛化能力。为克服这些局限，我们首次将元检测引入AI生成视频检测，通过在强化学习中联合优化预测标签与支撑证据，实现可靠的伪造检测。该范式需要可靠的证据信号及将其整合到标签级优化的有效机制。文本理由提供伪造痕迹的语义描述，但其生成与验证依赖外部模型，导致监督易受幻觉和语义偏差影响。相比之下，时间定位提供更客观、可验证的证据，因为伪造构建过程中可精确控制被篡改的时间区间。基于此洞见，我们提出自动化数据构建流水线，通过用边界帧条件视频生成模型替换时间片段，生成配对的真实-伪造视频。此外，我们引入证据引导的奖励再分配机制，该机制根据证据质量在标签正确的响应间重新分配奖励，在保留可靠标签监督的同时，鼓励检测器获取细粒度且可验证的伪造定位能力。大量实验表明，VidForensics-M1可有效利用可验证的时间证据，实现鲁棒且可泛化的AI生成视频检测。
英文摘要
Recent advances in video generation models have significantly improved the realism of synthetic videos, blurring the boundary between generated and authentic content and raising concerns about misinformation. Existing MLLM-based detectors mainly rely on supervised fine-tuning or label-level reinforcement learning, where coarse supervision limits generalization to unseen scenarios and emerging video generators. To overcome these limitations, we are the first to introduce \textbf{meta-detection} into AI-generated video detection, enabling reliable forgery detection by jointly optimizing predicted labels and supporting evidence within reinforcement learning. This paradigm requires reliable evidence signals and effective mechanisms to integrate them into label-level optimization. Textual rationales provide semantic descriptions of forgery artifacts, but their generation and verification depend on external models, making supervision vulnerable to hallucinations and semantic biases. In contrast, temporal grounding provides more objective and verifiable evidence, as manipulated intervals can be precisely controlled during forgery construction. Based on this insight, we propose an automated data construction pipeline that generates paired real-fake videos by replacing temporal segments with boundary-frame-conditioned video generation models. Furthermore, we introduce \textbf{Evidence-Guided Reward Redistribution}, which performs evidence-aware credit assignment by redistributing rewards among label-correct responses according to evidence quality. This preserves reliable label supervision while encouraging detectors to acquire fine-grained and verifiable forgery localization capabilities. Extensive experiments demonstrate that \textbf{VidForensics-M1} effectively leverages verifiable temporal evidence to achieve robust and generalizable AI-generated video detection.
Comments27 pages, 15 figures
