| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Knowledge-Based Visual Question Answering (Direct Answer) | A-OK-VQA (test) | Accuracy57.5 | 11 | |
| Knowledge-Based Visual Question Answering (Direct Answer) | A-OK-VQA (val) | Accuracy0.586 | 10 | |
| Knowledge-Based Visual Question Answering (Multiple Choice) | A-OK-VQA (test) | Accuracy57.3 | 6 | |
| Knowledge-Based Visual Question Answering (Multiple Choice) | A-OK-VQA (val) | Accuracy60.3 | 6 |