# Human Activity Evaluation on SceneTeract (test) Source: [Human Activity Evaluation on SceneTeract (test)]() Collection index: [llms.txt]() Data updated: 2026-04-01T08:44:28.899Z Evaluation of Human Activity Evaluation performance on the test split of the SceneTeract dataset, reporting metrics including Accuracy, FP, MCC, InGap, HSI, and Cons. 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 | Human Activity Evaluation | | Dataset (including any stated split) | SceneTeract (test) | | 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. - [FP](): Lower is better; unit: SCALAR. - [MCC](): Higher is better; unit: SCALAR. - [InGap](): Lower is better; unit: SCALAR. - [HSI](): Higher is better; unit: SCALAR. - [Consistency](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Gemini-3-Flash-Preview | 69.9 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-4B-Instruct with GRPO | 69.2 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Claude-Sonnet-4-6 | 62.8 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Claude-Sonnet-4-6 | 62.3 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-8B-Instruct | 62.3 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemini-3-Flash-Preview | 62.1 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemini-3.1-Pro-Preview | 61.7 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemma3-12B-Instruct | 61.7 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemini-3.1-Pro-Preview | 61.6 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-8B-Instruct | 61.5 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-4B-Instruct with GRPO | 61.5 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-4B-Instruct | 61.4 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Qwen3-VL-4B-Instruct | 60.8 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemma3-4B-Instruct | 60.1 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemma3-4B-Instruct | 59.8 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Gemma3-12B-Instruct | 59.3 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Ministral3-3B-Instruct | 54 | Not recorded | 69cca6a10636e5bf4afcbc36 | | Ministral3-3B-Instruct | 41.7 | Not recorded | 69cca6a10636e5bf4afcbc36 | ### Gemini-3-Flash-Preview Accuracy: 69.9. Result ID: 69ccdad947f34ce924b69eb1. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-Instruct with GRPO Accuracy: 69.2. Result ID: 69ccdad947f34ce924b69ec1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Claude-Sonnet-4-6 Accuracy: 62.8. Result ID: 69ccdad947f34ce924b69eb4. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Claude-Sonnet-4-6 Accuracy: 62.3. Result ID: 69ccdad947f34ce924b69eb5. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-Instruct Accuracy: 62.3. Result ID: 69ccdad947f34ce924b69eb7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Gemini-3-Flash-Preview Accuracy: 62.1. Result ID: 69ccdad947f34ce924b69eb0. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Gemini-3.1-Pro-Preview Accuracy: 61.7. Result ID: 69ccdad947f34ce924b69eb2. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Gemma3-12B-Instruct Accuracy: 61.7. Result ID: 69ccdad947f34ce924b69eb9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Gemini-3.1-Pro-Preview Accuracy: 61.6. Result ID: 69ccdad947f34ce924b69eb3. [Reporting paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Qwen3-VL-8B-Instruct Accuracy: 61.5. Result ID: 69ccdad947f34ce924b69eb6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-Instruct with GRPO Accuracy: 61.5. Result ID: 69ccdad947f34ce924b69ec0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-Instruct Accuracy: 61.4. Result ID: 69ccdad947f34ce924b69ebf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Qwen3-VL-4B-Instruct Accuracy: 60.8. Result ID: 69ccdad947f34ce924b69ebe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Gemma3-4B-Instruct Accuracy: 60.1. Result ID: 69ccdad947f34ce924b69ebd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. ### Gemma3-4B-Instruct Accuracy: 59.8. Result ID: 69ccdad947f34ce924b69ebc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Gemma3-12B-Instruct Accuracy: 59.3. Result ID: 69ccdad947f34ce924b69eb8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Ministral3-3B-Instruct Accuracy: 54. Result ID: 69ccdad947f34ce924b69eba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Direct"}. This record does not assert independent reproduction. ### Ministral3-3B-Instruct Accuracy: 41.7. Result ID: 69ccdad947f34ce924b69ebb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Evaluation Setting":"Decomposed"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.