| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Grounded Situation Recognition | SWiG (dev) | Value Accuracy76.17 | 51 | |
| Grounded Situation Recognition | SWiG (test) | Value Accuracy75.95 | 33 | |
| Grounded Situation Recognition | SWiG v1 (dev) | Top-1 Predicted Verb Accuracy58.19 | 21 | |
| Grounded Situation Recognition | SWiG 1.0 (test) | Top-1 Verb Acc58.19 | 13 | |
| Human-Object Interaction Detection | SWIG HOI (test) | mAP (Non-rare)23.67 | 7 | |
| Situation Recognition | Swig (test-std) | Accuracy65 | 5 | |
| Human-Object Interaction Detection | SWiG | mAP (Full)16.21 | 3 |