Visual Reasoning On Nlvr2 Dev
Metrics
Accuracy
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Accuracy |
|---|---|
| toward-building-general-foundation-models-for | 87.6 |
| vlmo-unified-vision-language-pre-training | 85.64 |
| visualbert-a-simple-and-performant-baseline | 66.7 |
| differentiable-outlier-detection-enable | 83.9 |
| coca-contrastive-captioners-are-image-text | 86.1 |
| multi-grained-vision-language-pre-training | 84.41 |
| x-2-vlm-all-in-one-pre-trained-model-for | 88.7 |
| seeing-out-of-the-box-end-to-end-pre-training | 76.37 |
| align-before-fuse-vision-and-language | 83.14 |
| implicit-differentiable-outlier-detection | 84.6 |
| vilt-vision-and-language-transformer-without | 75.7 |
| simvlm-simple-visual-language-model | 84.53 |
| image-as-a-foreign-language-beit-pretraining | 91.51 |
| x-2-vlm-all-in-one-pre-trained-model-for | 86.2 |
| lxmert-learning-cross-modality-encoder | 74.9 |