# Video Understanding on HERBench Lite Source: [Video Understanding on HERBench Lite]() Collection index: [llms.txt]() Data updated: 2026-03-20T09:19:25.004Z Evaluation of general video understanding capabilities using the HERBench Lite dataset, measured primarily by Accuracy. 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 Understanding | | Dataset (including any stated split) | HERBench Lite | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | VideoChat-A1 | 66.7 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 43.22 | Not recorded | 69bcd4760636e5bf4afc296a | | BOLT | 42.2 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform Sampling | 41.7 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 40.68 | Not recorded | 69bcd4760636e5bf4afc296a | | AKS | 40.25 | Not recorded | 69bcd4760636e5bf4afc296a | | T\* | 39.1 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 38.32 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 38.3 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 37.68 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 37.47 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 37.27 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 35.87 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 35.75 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 35.17 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 34.05 | Not recorded | 69bcd4760636e5bf4afc296a | | HiMu | 31.47 | Not recorded | 69bcd4760636e5bf4afc296a | | Uniform | 31.2 | Not recorded | 69bcd4760636e5bf4afc296a | ### VideoChat-A1 Accuracy: 66.7. Result ID: 69bd110b47f34ce924a7601a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"GPT-4o","Frame Budget (K)":"384","Evaluation Protocol":"Stratified random 25% subset"}. This record does not assert independent reproduction. ### HiMu Accuracy: 43.22. Result ID: 69bd110b47f34ce924a7600e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### BOLT Accuracy: 42.2. Result ID: 69bd110b47f34ce924a7600b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### Uniform Sampling Accuracy: 41.7. Result ID: 69bd110b47f34ce924a7600a. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### HiMu Accuracy: 40.68. Result ID: 69bd110b47f34ce924a7601b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"GPT-4o","Frame Budget (K)":"16","Evaluation Protocol":"Stratified random 25% subset"}. This record does not assert independent reproduction. ### AKS Accuracy: 40.25. Result ID: 69bd110b47f34ce924a7600d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### T\* Accuracy: 39.1. Result ID: 69bd110b47f34ce924a7600c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### HiMu Accuracy: 38.32. Result ID: 69bd110b47f34ce924a76012. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"InternVL-3.5-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### Uniform Accuracy: 38.3. Result ID: 69bd110b47f34ce924a76011. [Reporting paper]() Recorded evaluation configuration: {"Model":"InternVL-3.5-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### HiMu Accuracy: 37.68. Result ID: 69bd110b47f34ce924a76018. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Flash","Frame Budget (K)":"16","Evaluation Protocol":"Stratified random 25% subset"}. This record does not assert independent reproduction. ### Uniform Accuracy: 37.47. Result ID: 69bd110b47f34ce924a76019. [Reporting paper]() Recorded evaluation configuration: {"Model":"GPT-4o","Frame Budget (K)":"16","Evaluation Protocol":"Stratified random 25% subset"}. This record does not assert independent reproduction. ### Uniform Accuracy: 37.27. Result ID: 69bd110b47f34ce924a76017. [Reporting paper]() Recorded evaluation configuration: {"Model":"Gemini-2.5-Flash","Frame Budget (K)":"16","Evaluation Protocol":"Stratified random 25% subset"}. This record does not assert independent reproduction. ### HiMu Accuracy: 35.87. Result ID: 69bd110b47f34ce924a76010. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LLaVA-OV-1.5-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### Uniform Accuracy: 35.75. Result ID: 69bd110b47f34ce924a7600f. [Reporting paper]() Recorded evaluation configuration: {"Model":"LLaVA-OV-1.5-8B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### HiMu Accuracy: 35.17. Result ID: 69bd110b47f34ce924a76014. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### Uniform Accuracy: 34.05. Result ID: 69bd110b47f34ce924a76013. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### HiMu Accuracy: 31.47. Result ID: 69bd110b47f34ce924a76016. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Gemma-3-12B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. ### Uniform Accuracy: 31.2. Result ID: 69bd110b47f34ce924a76015. [Reporting paper]() Recorded evaluation configuration: {"Model":"Gemma-3-12B","Frame Budget (K)":"16","Evaluation Protocol":"Full benchmark"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.