Cross-Modal Information Retrieval on CommercialAdsDataset
Metric: ADD(S) AUC (higher is better)
Metric: ADD(S) AUC (higher is better)
| # | Model↕ | ADD(S) AUC▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | AlignCMSS | 91.73 | No | Align before Search: Aligning Ads Image to Text ... | 2023-09-28 | Code |
| 2 | VinVL | 88.56 | No | VinVL: Revisiting Visual Representations in Visi... | 2021-01-02 | Code |
| 3 | AdsCVLR | 87.9 | No | - | - | - |
| 4 | OSCAR | 87.45 | No | Oscar: Object-Semantics Aligned Pre-training for... | 2020-04-13 | Code |
| 5 | VL-BERT | 86.27 | No | VL-BERT: Pre-training of Generic Visual-Linguist... | 2019-08-22 | Code |
| 6 | BLIP | 83.51 | No | BLIP: Bootstrapping Language-Image Pre-training ... | 2022-01-28 | Code |
| 7 | Unicoder-VL | 83.16 | No | Unicoder-VL: A Universal Encoder for Vision and ... | 2019-08-16 | - |
| 8 | ALBEF | 82.74 | No | Align before Fuse: Vision and Language Represent... | 2021-07-16 | Code |