Multi Label Text Classification On Cc3M
Métriques
Accuracy
F1
Precision
Recall
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | Accuracy | F1 | Precision | Recall | Paper Title | Repository |
|---|---|---|---|---|---|---|
| Qwen-72B | 80.9 | 62.1 | 69.3 | 56.2 | Qwen Technical Report | |
| Vicuna-33B | 75.9 | 60.4 | 52.7 | 70.7 | - | - |
| TTD (w/o fine-tuning) | 91.0 | 78.5 | 82.9 | 74.5 | TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias | |
| NLTK | 79.6 | 69.8 | 59.8 | 83.7 | NLTK: The Natural Language Toolkit | |
| Vicuna-7B | 70.9 | 54.4 | 44.1 | 71.0 | - | - |
| TTD (w/ fine-tuning) | 88.6 | 82.8 | 88.3 | 78.0 | TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias |
0 of 6 row(s) selected.