Are Multimodal LLMs Ready for Surveillance? A Reality Check on Zero-Shot Anomaly Detection in the Wild
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914578542100480 |
|---|---|
| author | Yao, Shanle Pazho, Armin Danesh Rashvand, Narges Tabkhi, Hamed |
| author_facet | Yao, Shanle Pazho, Armin Danesh Rashvand, Narges Tabkhi, Hamed |
| contents | Multimodal large language models (MLLMs) have demonstrated impressive general competence in video understanding, yet their reliability for real-world Video Anomaly Detection (VAD) remains largely unexplored. Unlike conventional pipelines relying on reconstruction or pose-based cues, MLLMs enable a paradigm shift: treating anomaly detection as a language-guided reasoning task. In this work, we systematically evaluate state-of-the-art MLLMs on the ShanghaiTech and CHAD benchmarks by reformulating VAD as a binary classification task under weak temporal supervision. We investigate how prompt specificity and temporal window lengths (1s--3s) influence performance, focusing on the precision--recall trade-off. Our findings reveal a pronounced conservative bias in zero-shot settings; while models exhibit high confidence, they disproportionately favor the 'normal' class, resulting in high precision but a recall collapse that limits practical utility. We demonstrate that class-specific instructions can significantly shift this decision boundary, improving the peak F1-score on ShanghaiTech from 0.09 to 0.64, yet recall remains a critical bottleneck. These results highlight a significant performance gap for MLLMs in noisy environments and provide a foundation for future work in recall-oriented prompting and model calibration for open-world surveillance, which demands complex video understanding and reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04727 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Are Multimodal LLMs Ready for Surveillance? A Reality Check on Zero-Shot Anomaly Detection in the Wild Yao, Shanle Pazho, Armin Danesh Rashvand, Narges Tabkhi, Hamed Computer Vision and Pattern Recognition Artificial Intelligence Multimodal large language models (MLLMs) have demonstrated impressive general competence in video understanding, yet their reliability for real-world Video Anomaly Detection (VAD) remains largely unexplored. Unlike conventional pipelines relying on reconstruction or pose-based cues, MLLMs enable a paradigm shift: treating anomaly detection as a language-guided reasoning task. In this work, we systematically evaluate state-of-the-art MLLMs on the ShanghaiTech and CHAD benchmarks by reformulating VAD as a binary classification task under weak temporal supervision. We investigate how prompt specificity and temporal window lengths (1s--3s) influence performance, focusing on the precision--recall trade-off. Our findings reveal a pronounced conservative bias in zero-shot settings; while models exhibit high confidence, they disproportionately favor the 'normal' class, resulting in high precision but a recall collapse that limits practical utility. We demonstrate that class-specific instructions can significantly shift this decision boundary, improving the peak F1-score on ShanghaiTech from 0.09 to 0.64, yet recall remains a critical bottleneck. These results highlight a significant performance gap for MLLMs in noisy environments and provide a foundation for future work in recall-oriented prompting and model calibration for open-world surveillance, which demands complex video understanding and reasoning. |
| title | Are Multimodal LLMs Ready for Surveillance? A Reality Check on Zero-Shot Anomaly Detection in the Wild |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2603.04727 |