# VAU-R1: Advancing Video Anomaly Understanding via Reinforcement Fine-Tuning Source: [VAU-R1: Advancing Video Anomaly Understanding via Reinforcement Fine-Tuning]() Collection index: [llms.txt]() Data updated: 2025-10-26T04:50:02.046Z Authors: Liyun Zhu, Qixiang Chen, Xi Shen, Xiaodong Cun Published: 2025-05-29T14:48:10.000Z ## Abstract Video Anomaly Understanding (VAU) is essential for applications such as smart cities, security surveillance, and disaster alert systems, yet remains challenging due to its demand for fine-grained spatio-temporal perception and robust reasoning under ambiguity. Despite advances in anomaly detection, existing methods often lack interpretability and struggle to capture the causal and contextual aspects of abnormal events. This limitation is further compounded by the absence of comprehensive benchmarks for evaluating reasoning ability in anomaly scenarios. To address both challenges, we introduce VAU-R1, a data-efficient framework built upon Multimodal Large Language Models (MLLMs), which enhances anomaly reasoning through Reinforcement Fine-Tuning (RFT). Besides, we propose VAU-Bench, the first Chain-of-Thought benchmark tailored for video anomaly reasoning, featuring multiple-choice QA, detailed rationales, temporal annotations, and descriptive captions. Empirical results show that VAU-R1 significantly improves question answering accuracy, temporal grounding, and reasoning coherence across diverse contexts. Together, our method and benchmark establish a strong foundation for interpretable and reasoning-aware video anomaly understanding. Our code is available at https://github.com/GVCLab/VAU-R1. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Video Anomaly Question Answering on MSAD](): Video Anomaly Question Answering; MSAD; Acc (w/o think): 88.33. - [Video Anomaly Question Answering on UCF-Crime](): Video Anomaly Question Answering; UCF-Crime; Accuracy (w/o think): 92.03. - [Video Anomaly Understanding Evaluation on UCF-Crime](): Video Anomaly Understanding Evaluation; UCF-Crime; CLS: 4.42. - [Video Question Answering on ECVA](): Video Question Answering; ECVA; Accuracy: 89.53. - [Video Anomaly Understanding Evaluation on MSAD](): Video Anomaly Understanding Evaluation; MSAD; CLS Score: 5.97. - [Video Anomaly Reasoning Evaluation on ECVA](): Video Anomaly Reasoning Evaluation; ECVA; CLS Score: 1.45. ## Original sources and implementation [Original paper]() [PDF]()