Image Text Matching On Commercialadsdataset
评估指标
ADD(S) AUC
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | ADD(S) AUC |
|---|---|
| oscar-object-semantics-aligned-pre-training | 87.45 |
| unicoder-vl-a-universal-encoder-for-vision | 83.16 |
| vinvl-making-visual-representations-matter-in | 88.56 |
| blip-bootstrapping-language-image-pre | 83.51 |
| align-before-search-aligning-ads-image-to | 91.73 |
| align-before-fuse-vision-and-language | 82.74 |
| vl-bert-pre-training-of-generic-visual | 86.27 |
| adscvlr-commercial-visual-linguistic | 87.90 |