MAMI
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Meme Classification | MAMI | Accuracy0.799 | 30 | |
| Harmful Meme Detection | MAMI | Accuracy85.5 | 19 | |
| Hateful Meme Detection | MAMI | F1 (Binary)82.3 | 17 | |
| Classification | MAMI | Accuracy89.1 | 16 | |
| Intervention | MAMI | rgL0.395 | 10 | |
| Explanation | MAMI | rgL0.238 | 10 | |
| Harmful meme detection | MAMI (test) | Accuracy81 | 10 | |
| Hateful Meme Detection | MAMI 1.0 (test) | AUC83.77 | 9 | |
| Multi-class Classification | MAMI Task B | Macro-F179.6 | 7 | |
| Misogyny Identification | MAMI | Accuracy80.1 | 6 |
Showing 10 of 10 rows