| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Video Question Answering | ActivityNet-QA | Accuracy64.4 | 319 | |
| Video Question Answering | ActivityNet-QA (test) | Accuracy82.78 | 275 | |
| Video Question Answering | ActivityNet-QA zero-shot (test) | Accuracy60.1 | 55 | |
| Video Question Answering | ActivityNet-QA LLaVA-Hound in-domain (test) | Accuracy68.5 | 11 | |
| Video Question Answering | ActivityNet-QA multi-event | VQA Accuracy62.21 | 6 | |
| Video Question Answering | ActivityNet-QA (val) | Accuracy44.1 | 6 | |
| Short Video Generative Performance Evaluation | ActivityNet-QA (test) | Info Correctness2.76 | 5 |