WeVisForensics: A New Method for AI Video Detection
The story so far
Hugging Face has released WeVisForensics, a self-evolving intelligent agent for full-modal AI-generated video detection. Through a dual-time-scale reinforcement learning strategy, it coordinates self-evolving visual and audio forensic inspection agents to detect generated video artifacts with high accuracy and provide local evidence. In domain, out-of-domain, and agent benchmark tests, WeVisForensics achieves an average detection accuracy of 95.2%, 9.0% higher than the strongest prior detector.
AI-written from the reports · Updated 1 hour ago
Coverage timeline
Follow the coverage to see every side of the story.
- Hugging Face Daily PapersWeVisForensics: Self-Evolving Agent for Omni AI-Generated Video Detection
WeVisForensics coordinates an autonomic visual and audio forensic inspection intelligent agent through a dual-time-scale reinforcement learning strategy to detect generated video artifacts with high accuracy and provide local evidence, achieving an average detection accuracy of 95.2% in domain, cross-domain, and agent benchmark tests, which is 9.0% higher than the strongest prior detector.
Heat over time
Not enough data yet to draw a trend.