Visual Commonsense Reasoning On Gd Vcr
평가 지표
Accuracy
Gap (West)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | Accuracy | Gap (West) | Paper Title | Repository |
|---|---|---|---|---|
| VisualBERT | 53.95 | -10.42 | Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning | |
| Human | 88.84 | - | Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning | |
| ViLBERT | 59.99 | -7.28 | Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning | |
| Text-only BERT | 35.33 | - | Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning |
0 of 4 row(s) selected.