Recent advances in foundation video generators such as Sora2, Veo3, and other commercial systems have produced highly realistic synthetic videos, exposing the limitations of existing detection methods that rely on shallow embedding trajectories, image-based adaptation, or computationally heavy MLLMs. We propose EA-Swin, an Embedding-Agnostic Swin Transformer that models spatiotemporal dependencies directly on pretrained video embeddings via a factorized windowed attention design, making it compatible with generic ViT-style patch-based encoders. Alongside the model, we construct the EA-Video dataset, a benchmark dataset comprising 130K videos that integrates newly collected samples with curated existing datasets, covering diverse commercial and open-source generators and including unseen-generator splits for rigorous cross-distribution evaluation. Extensive experiments show that EA-Swin achieves 0.97-0.99 accuracy across major generators, outperforming prior SoTA methods (typically 0.8-0.9) by a margin of 5-20%, while maintaining strong generalization to unseen distributions, establishing a scalable and robust solution for modern AI-generated video detection.

翻译：近期，以Sora2、Veo3及其他商业系统为代表的基础视频生成器取得了显著进展，能够生成高度逼真的合成视频，这暴露了现有检测方法的局限性：这些方法依赖于浅层嵌入轨迹、基于图像的适配或计算量巨大的多模态大语言模型。我们提出EA-Swin，一种嵌入无关的Swin Transformer，它通过因子化窗口注意力设计直接在预训练视频嵌入上建模时空依赖关系，使其能够兼容通用的ViT风格基于补丁的编码器。与此同时，我们构建了EA-Video数据集，这是一个包含13万个视频的基准数据集，整合了新收集的样本与精选的现有数据集，涵盖了多样化的商业及开源生成器，并包含针对未见生成器的数据划分以进行严格的跨分布评估。大量实验表明，EA-Swin在主流生成器上实现了0.97-0.99的准确率，以5-20%的优势超越了现有最佳方法（通常为0.8-0.9），同时在未见分布上保持了强大的泛化能力，为现代AI生成视频检测提供了一个可扩展且鲁棒的解决方案。