2026-11-02 · America/Los_Angeles · 论文 · #10

Venus-DefakerOne2: A Vision-Forensics Foundation Model with Broader Coverage

Real-world forgery is inherently heterogeneous: it manifests not only as digital manipulation but also as physical spoofing, and is carried not only by images but also by videos. Despite substantial progress, a unified anti-forgery foundation model covering such a broad threat spectrum remains absent. In this paper, we present Venus-DefakerOne2, which substantially extends the forensic capability boundary along both the forgery-form and media-type dimensions. Trained on tens of millions of open-source and in-house samples, it incorporates two key designs. Tide-Style-Tuning alternates LoRA…

热度 39.9 / 100；排名与评分保留该期记录。