Metric: B3 (higher is better)
| # | Model↕ | B3▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | GIT2 | 58.9 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 2 | GIT | 58.46 | No | GIT: A Generative Image-to-text Transformer for ... | 2022-05-27 | Code |
| 3 | VLAF2 | 51.1 | No | - | - | - |
| 4 | Microsoft Cognitive Services team | 48.73 | No | VIVO: Visual Vocabulary Pre-Training for Novel O... | 2020-09-28 | - |
| 5 | icp2ssi1_coco_si_0.02_5_test | 43.22 | No | - | - | - |
| 6 | test_cbs2 | 40.26 | No | - | - | - |
| 7 | UpDown + ELMo + CBS | 37.04 | No | - | - | - |
| 8 | UpDown | 36.91 | No | - | - | - |
| 9 | Human | 36.84 | No | - | - | - |
| 10 | Neural Baby Talk | 32.37 | No | - | - | - |
| 11 | Neural Baby Talk + CBS | 30.66 | No | - | - | - |