| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Video Question Answering | EgoTaskQA (normal) | Accuracy (World)47.32 | 12 | |
| Embodied Question Answering | EgoTaskQA (test) | Exact Match80 | 10 | |
| Video Question Answering | EgoTaskQA indirect | Accuracy (World)47.82 | 9 | |
| Embodied Cognition | EgoTaskQA | Score85 | 4 |