FHM
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Harmful meme detection | FHM (test) | Accuracy76.2 | 51 | |
| Harmful Meme Detection | FHM | Macro-F175.79 | 49 | |
| Binary Hate Detection | FHM (test) | Accuracy78.6 | 25 | |
| Hateful Meme Classification and Explanation | FHM | Accuracy77 | 18 | |
| Classification | FHM | Accuracy80.26 | 16 | |
| Hateful Meme Detection | FHM | AUC80.87 | 12 | |
| Intervention | FHM | rgL29 | 10 | |
| Explanation | FHM | rgL0.242 | 10 | |
| Hateful Meme Detection | FHM 1.0 (test) | AUC83.58 | 9 | |
| Harm classification | FHM | Accuracy74.1 | 7 | |
| Interpretability Evaluation | FHM (test) | Faithfulness7.56 | 2 |
Showing 11 of 11 rows