| image-captioning-on-coco-captions | SimVLM | #16 | BLEU-4: 40.6CIDER: 143.3METEOR: 33.4SPICE: 25.4 |
| image-captioning-on-nocaps-entire | Single Model | #3 | CIDEr: 110.31B1: 83.78B2: 68.86B3: 51.06B4: 32.2… |
| image-captioning-on-nocaps-in-domain | Single Model | #5 | CIDEr: 108.98B1: 84.64B2: 70.0B3: 52.96B4: 34.66… |
| image-captioning-on-nocaps-near-domain | Single Model | #4 | CIDEr: 110.76B1: 84.36B2: 69.83B3: 52.42B4: 33.74… |
| image-captioning-on-nocaps-out-of-domain | Single Model | #4 | CIDEr: 109.49B1: 80.89B2: 64.21B3: 44.38B4: 24.47… |
| image-captioning-on-nocaps-val-in-domain | SimVLM | #5 | CIDEr: 113.7Pre-train (#images): 1.8B |
| image-captioning-on-nocaps-val-near-domain | SimVLM | #5 | CIDEr: 110.9Pre-train (#images): 1.8B |
| image-captioning-on-nocaps-val-out-domain | SimVLM | #5 | CIDEr: 115.2Pretrain (#images): 1.8B |
| image-captioning-on-nocaps-val-overall | SimVLM | #5 | CIDEr: 112.2Pretrain (#images): 1.8B |
| visual-entailment-on-snli-ve-test | SimVLM | #4 | Accuracy: 86.32 |
| visual-entailment-on-snli-ve-val | SimVLM | #4 | Accuracy: 86.21 |
| visual-question-answering-on-vqa-v2-test-dev | SimVLM | #11 | Accuracy: 80.03 |
| visual-question-answering-on-vqa-v2-test-std | SimVLM | #6 | overall: 80.34 |
| visual-reasoning-on-nlvr2-dev | SimVLM | #8 | Accuracy: 84.53 |
| visual-reasoning-on-nlvr2-test | SimVLM | #7 | Accuracy: 85.15 |