# Video Anomaly Reasoning on Video Anomaly Reasoning (test) Source: [Video Anomaly Reasoning on Video Anomaly Reasoning (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T06:00:47.562Z Evaluation of Video Anomaly Reasoning performance on the test split of the Video Anomaly Reasoning dataset, measured using specific performance indicators (RR, RW, WR, WW, RRR, WWW). 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. | Scope | Recorded value | | --- | --- | | Task | Video Anomaly Reasoning | | Dataset (including any stated split) | Video Anomaly Reasoning (test) | | Metric | RR Score | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 27 of 27 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [RR Score](): Higher is better; unit: SCALAR. - [RW Score](): Higher is better; unit: SCALAR. - [WR Score](): Higher is better; unit: SCALAR. - [WW Score](): Higher is better; unit: SCALAR. - [RRR Score](): Higher is better; unit: SCALAR. - [WWW Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | RR Score | Result date | Report paper | | --- | --- | --- | --- | | Vad-R1-Plus | 0.571 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Grok4-Fast | 0.36 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Gemini2.5-Flash | 0.298 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Vad-R1 | 0.286 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Qwen3VL-Thinking | 0.247 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | GPT5-mini | 0.219 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Qwen3VL-Instruct | 0.206 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Video-R1 | 0.195 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | VideoChat-R1 | 0.177 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Kimi-VL-Instruct | 0.172 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Kimi-VL-Thinking | 0.093 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | InternVL3.5 | 0.069 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | MiniCPM-V4.5 | 0.057 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | OpenR1-Video | 0.055 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | InternVideo2.5 | 0.031 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | MiMo-VL | 0.021 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Ovis2.5 | 0.021 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Holmes-VAD | 0.017 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | HAWK | 0.014 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Keye-VL1.5 | 0.012 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | LLaVA-NeXT-Video | 0.009 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | VideoChat-Flash | 0.008 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | LLaVA-OneVision | 0.007 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Holmes-VAU | 0.007 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | VideoLLaMA3 | 0.006 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | Video-XL | 0.002 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | | GLM4.1-V | 0.001 | Not recorded | 6969c5e7e25ad2fa9d1ec759 | ### Vad-R1-Plus RR Score: 0.571. Result ID: 6997b9a85698665ec4102824. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Ours"}. This record does not assert independent reproduction. ### Grok4-Fast RR Score: 0.36. Result ID: 6997b9a85698665ec4102822. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Proprietary MLLMs"}. This record does not assert independent reproduction. ### Gemini2.5-Flash RR Score: 0.298. Result ID: 6997b9a85698665ec4102820. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Proprietary MLLMs"}. This record does not assert independent reproduction. ### Vad-R1 RR Score: 0.286. Result ID: 6997b9a85698665ec4102823. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Ours","Status":"Preliminary"}. This record does not assert independent reproduction. ### Qwen3VL-Thinking RR Score: 0.247. Result ID: 6997b9a85698665ec4102816. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### GPT5-mini RR Score: 0.219. Result ID: 6997b9a85698665ec4102821. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Proprietary MLLMs"}. This record does not assert independent reproduction. ### Qwen3VL-Instruct RR Score: 0.206. Result ID: 6997b9a85698665ec4102810. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### Video-R1 RR Score: 0.195. Result ID: 6997b9a85698665ec4102814. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### VideoChat-R1 RR Score: 0.177. Result ID: 6997b9a85698665ec4102815. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### Kimi-VL-Instruct RR Score: 0.172. Result ID: 6997b9a85698665ec4102811. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### Kimi-VL-Thinking RR Score: 0.093. Result ID: 6997b9a85698665ec410281a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### InternVL3.5 RR Score: 0.069. Result ID: 6997b9a85698665ec410280b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### MiniCPM-V4.5 RR Score: 0.057. Result ID: 6997b9a85698665ec4102819. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### OpenR1-Video RR Score: 0.055. Result ID: 6997b9a85698665ec4102813. [Reporting paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### InternVideo2.5 RR Score: 0.031. Result ID: 6997b9a85698665ec410280a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### MiMo-VL RR Score: 0.021. Result ID: 6997b9a85698665ec4102818. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### Ovis2.5 RR Score: 0.021. Result ID: 6997b9a85698665ec410281b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### Holmes-VAD RR Score: 0.017. Result ID: 6997b9a85698665ec410281d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"MLLMs-based VAD Methods"}. This record does not assert independent reproduction. ### HAWK RR Score: 0.014. Result ID: 6997b9a85698665ec410281f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"MLLMs-based VAD Methods"}. This record does not assert independent reproduction. ### Keye-VL1.5 RR Score: 0.012. Result ID: 6997b9a85698665ec4102817. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. ### LLaVA-NeXT-Video RR Score: 0.009. Result ID: 6997b9a85698665ec410280e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### VideoChat-Flash RR Score: 0.008. Result ID: 6997b9a85698665ec410280c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### LLaVA-OneVision RR Score: 0.007. Result ID: 6997b9a85698665ec410280f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### Holmes-VAU RR Score: 0.007. Result ID: 6997b9a85698665ec410281e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model Category":"MLLMs-based VAD Methods"}. This record does not assert independent reproduction. ### VideoLLaMA3 RR Score: 0.006. Result ID: 6997b9a85698665ec410280d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### Video-XL RR Score: 0.002. Result ID: 6997b9a85698665ec4102812. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video MLLMs"}. This record does not assert independent reproduction. ### GLM4.1-V RR Score: 0.001. Result ID: 6997b9a85698665ec410281c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Category":"Video Reasoning MLLMs"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.