Visual Question Answering On Vcr Q Ar Test
Métriques
Accuracy
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Accuracy |
|---|---|
| kvl-bert-knowledge-enhanced-visual-and | 60.3 |
| unifying-vision-and-language-tasks-via-text | 58.9 |
| ernie-vil-knowledge-enhanced-vision-language | 70.5 |
| uniter-learning-universal-image-text-1 | 62.8 |
| gpt4roi-instruction-tuning-large-language | 81.6 |
| visualbert-a-simple-and-performant-baseline | 52.4 |
| vl-bert-pre-training-of-generic-visual | 59.7 |