NVIDIA's ReactVAU: Slow-Fast Decoupled Framework for Real-Time Streaming Video Anomaly Understanding
nvidia · hf · 2026-09-09
NVIDIA released ReactVAU, a slow-fast decoupled framework for real-time streaming video anomaly understanding.
- A fast detection module handles real-time anomaly detection with persistent anomaly-aware memory
- Heavy reasoning runs on-demand via a slow module, minimizing expensive model usage
- Designed to make streaming anomaly understanding both real-time and compute-efficient
More from Research
- Stanford Data-Centric AI Workshop reopens site, extends poster deadline — jure · 2026-10-26
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Ito: a browser that detects browsing intent and auto-organizes tabs into Flows — andrewkuznet · 2026-10-26
- Cornell study says brains process language as linear sequences, challenging Chomsky's hierarchy — anirbanbandyo · 2026-10-26
- Hi3D V1.0 open-sourced with 2M training assets and 90% slicing success rate — kimmonismus · 2026-10-26
- Ortet evaluates 34 open-weight models on 50 health tasks, finds general leaderboards insufficient — elmanmansimov · 2026-10-26