ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding

Hugging Face Daily Papers Papers
video-anomaly-detection streaming-video slow-fast-framework real-time-inference memory-persistence mllm spatial-grid-folding

Summary

ReactVAU introduces a slow-fast decoupled framework for real-time streaming video anomaly understanding, leveraging a fast detection module, persistent anomaly-aware memory, and on-demand slow reasoning to enhance efficiency and performance.

In this paper, we propose ReactVAU, a Slow-Fast Decoupled Framework for real-time streaming Video Anomaly Understanding (VAU). Existing VAU methods rely on offline inference with global temporal sampling, which violates causality and prevents deployment in live surveillance streams. Conversely, general streaming video models satisfy causal access but dilute rare transient anomalies during memory compression and often invoke heavyweight MLLMs uniformly over long normal intervals. React VAU addresses this gap with three synergistic components: a lightweight Fast Detection Module based on Spatial Grid Folding (SGF) for continuous anomaly filtering; an Anomaly-Aware Persistent Memory (AAPM) that protects critical visual cues from temporal decay; and a heavyweight Slow Reasoning Module that remains dormant during normal streams and is awakened only by suspicious events for semantic verification and causal description. Extensive experiments on multiple benchmarks demonstrate that ReactVAU operates under strict streaming constraints while simultaneously achieving competitive performance in both anomaly detection and causal reasoning, alongside significantly enhanced computational efficiency by minimizing heavyweight MLLM invocations. Project page is available at https://huiyuiui.github.io/React_VAU/
Original Article
View Cached Full Text

Cached at: 09/09/26, 08:31 AM

Paper page - ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding

Source: https://huggingface.co/papers/2609.07941

Abstract

ReactVAU enables real-time streaming video anomaly understanding via a fast detection module, persistent anomaly-aware memory, and an on-demand slow reasoning module that minimizes heavy model usage.

In this paper, we propose ReactVAU, aSlow-Fast Decoupled Frameworkfor real-timestreaming VideoAnomaly Understanding (VAU). Existing VAU methods rely on offline inference with global temporal sampling, which violates causality and prevents deployment in live surveillance streams. Conversely, generalstreaming videomodels satisfy causal access but dilute rare transient anomalies during memory compression and often invoke heavyweightMLLMs uniformly over long normal intervals. React VAU addresses this gap with three synergistic components: a lightweight Fast Detection Module based onSpatial Grid Folding(SGF) for continuous anomaly filtering; anAnomaly-Aware Persistent Memory(AAPM) that protects critical visual cues from temporal decay; and a heavyweight Slow Reasoning Module that remains dormant during normal streams and is awakened only by suspicious events for semantic verification and causal description. Extensive experiments on multiple benchmarks demonstrate that ReactVAU operates under strict streaming constraints while simultaneously achieving competitive performance in both anomaly detection andcausal reasoning, alongside significantly enhanced computational efficiency by minimizing heavyweightMLLMinvocations. Project page is available at https://huiyuiui.github.io/React_VAU/

View arXiv pageView PDFProject pageGitHub2Add to collection

Get this paper in your agent:

hf papers read 2609\.07941

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.07941 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.07941 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.07941 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

V-RAE: Rethinking Video Latent Spaces for Generation

Hugging Face Daily Papers

V-RAE proposes a video representation autoencoder that builds semantically organized latents from frozen vision representations to enhance video generation quality, convergence speed, and predictive modeling.