{"ID":5937863,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-09T02:12:29.878472658Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.04607","arxiv_id":"2607.04607","title":"G2VD: Generalizable AI-Generated Video Detection via Counterfactual Intervention and Causal Disentanglement","abstract":"The rapid advancement of AI-generated videos poses increasing security risks and calls for robust detectors with strong cross-domain generalization. Although existing methods achieve promising results under in-domain evaluation, their performance often degrades substantially when tested on unseen generators. A key reason is shortcut learning, where detectors rely on domain-specific spurious cues, such as generator-dependent fingerprints and generation styles, instead of intrinsic forgery traces. To address this issue, we propose G2VD, a Generalizable AI-Generated Video Detection framework based on counterfactual intervention and causal disentanglement. First, G2VD introduces a counterfactual intervention pipeline (CFIPipeline) that generates controlled counterfactual samples via variational autoencoders (VAEs), followed by frequency-domain and pixel-domain alignment, thereby encouraging the detector to focus on generator-intrinsic cues. Building on this intervention process, we further design a causal disentanglement classifier consisting of two domain-anchored branches with distinct classification objectives, combined with an HSIC-based independence constraint to encourage the separation of task-relevant cues from domain-specific bias. Across four public datasets, G2VD shows strong average cross-domain performance and consistent gains over matched backbones. On the challenging GenVidBench cross-domain setting, it exceeds 90% accuracy and reaches an AUC close to 0.95. Notably, this performance is obtained using only 10% of the original training data. The code is available at https://github.com/dumeng98/G2VD.","short_abstract":"The rapid advancement of AI-generated videos poses increasing security risks and calls for robust detectors with strong cross-domain generalization. Although existing methods achieve promising results under in-domain evaluation, their performance often degrades substantially when tested on unseen generators. A key reas...","url_abs":"https://arxiv.org/abs/2607.04607","url_pdf":"https://arxiv.org/pdf/2607.04607v1","authors":"[\"Meng Du\",\"Hongchang Chen\",\"Ran Li\",\"Junjie Zhang\",\"Qi Ouyang\",\"Shuxin Liu\"]","published":"2026-07-06T02:25:18Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Variational Autoencoder\"]","has_code":false,"code_links":[{"ID":613996,"CreatedAt":"2026-07-07T03:14:33.014478982Z","UpdatedAt":"2026-07-07T03:14:33.014478982Z","DeletedAt":null,"paper_id":5937863,"paper_url":"https://arxiv.org/abs/2607.04607","paper_title":"G2VD: Generalizable AI-Generated Video Detection via Counterfactual Intervention and Causal Disentanglement","repo_url":"https://github.com/dumeng98/G2VD","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}