Hateful Meme Classification On Pridemm
Metrics
Accuracy
F1
Results
Performance results of various models on this benchmark
Model Name | Accuracy | F1 | Paper Title | Repository |
|---|---|---|---|---|
| CLIP (fine-tuned) | 72.4 | 72.3 | Learning Transferable Visual Models From Natural Language Supervision | - |
| LMM-RGCL (Qwen2-VL-2B) | 76.0 | 76.7 | Robust Adaptation of Large Multimodal Models for Retrieval Augmented Hateful Meme Detection | - |
| LMM-RGCL (Qwen2-VL-7B) | 78.1 | 78.4 | Robust Adaptation of Large Multimodal Models for Retrieval Augmented Hateful Meme Detection | - |
| MemeCLIP | 76.1 | 75.1 | MemeCLIP: Leveraging CLIP Representations for Multimodal Meme Classification | - |
| HateCLIPper | 75.5 | 74.1 | Hate-CLIPper: Multimodal Hateful Meme Classification based on Cross-modal Interaction of CLIP Features | - |
| RGCL | 76.3 | 76.5 | Improving Hateful Meme Detection through Retrieval-Guided Contrastive Learning | - |
0 of 6 row(s) selected.