SOURCE-LINKED INTELLIGENCE

NOVA: Normal-Side Modeling for Training-Free Zero-Shot Video Anomaly Detection

arXiv · AI, language, vision and robotics · article · Sep 6, 2026 · UTC

Training-free zero-shot video anomaly detection (ZS-VAD) leverages vision-language models (VLMs) to localize anomaly instances from a predefined anomaly vocabulary, without providing any video. Existing CLIP-based methods often emphasize anomaly-side semantics, while the competing normality side remains less carefully formulated. We identify two key limitations in existing solutions: (i) blurred decision boundary: normal prompts may contain ambiguous verbs, such as running, that are semantically close to anomalies, reducing normal and abnormal separation in the VLM embedding space; and (ii) mo

Read original source ↗ Open in workspace

recordType

paper

region

Global