Visual Question Answering Vqa On Imagenet
评估指标
ClipMatch@1
ClipMatch@5
Contains
ExactMatch
Follow-up ClipMatch@1
Follow-up ClipMatch@5
Follow-up Contains
Follow-up ExactMatch
评测结果
各个模型在此基准测试上的表现结果
模型名称 | ClipMatch@1 | ClipMatch@5 | Contains | ExactMatch | Follow-up ClipMatch@1 | Follow-up ClipMatch@5 | Follow-up Contains | Follow-up ExactMatch | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|---|---|
| BLIP-2 OPT | 57.10 | 77.24 | 35.49 | 0.87 | 67.22 | 83.54 | 40.31 | 2.54 | Open-ended VQA benchmarking of Vision-Language models by exploiting Classification datasets and their semantic hierarchy |
0 of 1 row(s) selected.