Visual Question Answering (VQA) on VizWiz 2020 Answerability
Metric: average_precision (higher is better)
Metric: average_precision (higher is better)
| # | Model↕ | average_precision▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | CLIP-Ensemble | 84.13 | No | Less Is More: Linear Layers on CLIP Features as ... | 2022-06-10 | - |
| 2 | CLIP-Single | 82.86 | No | Less Is More: Linear Layers on CLIP Features as ... | 2022-06-10 | - |
| 3 | VT-Transformer (MUL) | 76.96 | No | - | - | - |
| 4 | VT-Transformer (CAT) | 74.91 | No | - | - | - |
| 5 | BERT-RG-Regression | 52.22 | No | - | - | - |
| 6 | VWTest1 | 26.84 | No | - | - | - |