Image Captioning on nocaps XD (test)
102.4CIDErP2C
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| P2CPre-training data=CC3M2022.09 | 102.4 | 14.7 | |
| SimVLM_basePre-training data=1.8B2022.09 | 94.8 | 13.1 | |
| P2CPre-training data=COCO Caption 0.5M2022.09 | 93.5 | 14.1 | |
| Encoder-DecoderPre-training data=CC3M+CC12M2022.09 | 87.3 | 12 | |
| Encoder-DecoderPre-training data=CC12M [43]2022.09 | 85.3 | 11.8 |