Video Question Answering on TGIF-Frame (test)
75.6AccuracyMuLTI-L
Evaluation Results
| Method | Links | |
|---|---|---|
| MuLTI-L#PT=5.5M2023.03 | 75.6 | |
| mPLUG-2Pre-training Data Size=17M2023.02 | 75.4 | |
| GIT2Pre-training Data Size=12.9B2023.02 | 74.9 | |
| GIT22022.05 | 74.9 | |
| LAVENDERPre-training Data Size=30M2023.02 | 73.5 | |
| MuLTI-B#PT=5.5M2023.03 | 73.5 | |
| HiTeAPre-training Data Size=17M2023.02 | 73.2 | |
| GITPre-training Data Size=800M2023.02 | 72.8 | |
| GIT2022.05 | 72.8 | |
| GIT2022.05 | 72.8 | |
| mPLUG-2BasePre-training Data Size=17M2023.02 | 72.6 | |
| InternVideoPre-training Data Size=12M2023.02 | 72.2 | |
| Clover#PT=5.5M2023.03 | 71.6 | |
| MuLTI-S#PT=5.5M2023.03 | 71.2 | |
| MERLOTPre-training Data Size=180M2023.02 | 69.5 | |
| MERLOT2022.05 | 69.5 | |
| Prior SOTA2022.05 | 69.5 | |
| VIOLET#PT=185.5M2023.03 | 68.9 | |
| VIOLET2022.05 | 68.9 | |
| FrozenBiLMPre-training Data Size=10M2023.02 | 68.6 | |
| FrozenBiLM#PT=10M2023.03 | 68.6 | |
| All-in-onePre-training Data Size=283M2023.02 | 66.3 | |
| All-in-one2022.05 | 66.3 | |
| AllInOne#PT=102.5M2023.03 | 64.2 | |
| ClipBERT2022.05 | 60.3 | |
| QueST2022.05 | 59.7 | |
| HCRN2022.05 | 55.9 |