Visual Question Answering on OpenViVQA (test)
3.2129CIDErViTextBLIP-2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ViTextBLIP-2Version=base, Method Type=ViT-LM2024.04 | 3.2129 | 0.4717 | |
| QuMLPAGVersion=base, Method Type=CNN-LM2024.04 | 1.7082 | 0.2651 | |
| MLPAGVersion=base, Method Type=CNN-LM2024.04 | 1.6104 | 0.2739 | |
| M4CVersion=base, Method Type=CNN-LM2024.04 | 1.5073 | 0.2941 | |
| MCANVersion=base, Method Type=ViT-LM2024.04 | 1.0613 | 0.1699 | |
| LoRRAVersion=base, Method Type=CNN-RNN2024.04 | 0.8005 | 0.1349 | |
| FSTVersion=base, Method Type=CNN-LM2024.04 | 0.6141 | 0.105 |