{"ID":2891299,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.17394","arxiv_id":"2507.17394","title":"HiProbe-VAD: Video Anomaly Detection via Hidden States Probing in Tuning-Free Multimodal LLMs","abstract":"Video Anomaly Detection (VAD) aims to identify and locate deviations from normal patterns in video sequences. Traditional methods often struggle with substantial computational demands and a reliance on extensive labeled datasets, thereby restricting their practical applicability. To address these constraints, we propose HiProbe-VAD, a novel framework that leverages pre-trained Multimodal Large Language Models (MLLMs) for VAD without requiring fine-tuning. In this paper, we discover that the intermediate hidden states of MLLMs contain information-rich representations, exhibiting higher sensitivity and linear separability for anomalies compared to the output layer. To capitalize on this, we propose a Dynamic Layer Saliency Probing (DLSP) mechanism that intelligently identifies and extracts the most informative hidden states from the optimal intermediate layer during the MLLMs reasoning. Then a lightweight anomaly scorer and temporal localization module efficiently detects anomalies using these extracted hidden states and finally generate explanations. Experiments on the UCF-Crime and XD-Violence datasets demonstrate that HiProbe-VAD outperforms existing training-free and most traditional approaches. Furthermore, our framework exhibits remarkable cross-model generalization capabilities in different MLLMs without any tuning, unlocking the potential of pre-trained MLLMs for video anomaly detection and paving the way for more practical and scalable solutions.","short_abstract":"Video Anomaly Detection (VAD) aims to identify and locate deviations from normal patterns in video sequences. Traditional methods often struggle with substantial computational demands and a reliance on extensive labeled datasets, thereby restricting their practical applicability. To address these constraints, we propos...","url_abs":"https://arxiv.org/abs/2507.17394","url_pdf":"https://arxiv.org/pdf/2507.17394v1","authors":"[\"Zhaolin Cai\",\"Fan Li\",\"Ziwei Zheng\",\"Yanjun Qin\"]","published":"2025-07-23T10:41:46Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}