Visual Question Answering on VQA v2 (std)
84.3AccuracyPALI (16.9B)
Evaluation Results
| Method | Links | |
|---|---|---|
| PALI (16.9B)#Example=1.6B2023.04 | 84.3 | |
| CoCa#Example=4.8B2023.04 | 82.3 | |
| Flamingo (80B)#Example=2.3B2023.04 | 82.1 | |
| GIT2 (5.1B)#Example=12.9B2023.04 | 81.9 | |
| Florence#Example=900M2023.04 | 80.36 | |
| SimVLM#Example=1.8B2023.04 | 80.34 | |
| GIT#Example=800M2023.04 | 78.8 | |
| METERData=16M+, EPIC=true2022.11 | 78.7 | |
| VALORL#Example=33.5M, Resolution=3922023.04 | 78.62 | |
| X-VLMData=4M+, EPIC=true2022.11 | 78.5 | |
| METERData=16M+, EPIC=false2022.11 | 78.4 | |
| BLIP#Example=135M2023.04 | 78.32 | |
| X-VLMData=16M+, EPIC=true2022.11 | 78.3 | |
| X-VLMData=4M+, EPIC=false2022.11 | 78.2 | |
| X-VLMData=16M+, EPIC=false2022.11 | 78.2 | |
| METERData=4M+, EPIC=true2022.11 | 78 | |
| METERData=4M+, EPIC=false2022.11 | 77.9 | |
| METER#Example=10M2023.04 | 77.64 | |
| Shikra-13Bmode=zero-shot2023.08 | 77.5 | |
| UFO#Example=10M2023.04 | 76.76 | |
| ALBEFData=16M+, EPIC=true2022.11 | 76.7 | |
| VinVL#Example=10M2023.04 | 76.6 | |
| ALBEF#Example=20M2023.04 | 76.04 | |
| ALBEFData=16M+, EPIC=false2022.11 | 76 | |
| Ours-7B-[mix]mode=zero-shot2023.08 | 75.3 | |
| ALBEFData=4M+, EPIC=true2022.11 | 75.2 | |
| ALBEFData=4M+, EPIC=false2022.11 | 74.6 | |
| UNITER#Example=10M2023.04 | 74.02 | |
| Oscar#Example=10M2023.04 | 73.82 | |
| ViLTData=4M, EPIC=true2022.11 | 71.8 | |
| ViLTData=4M, EPIC=false2022.11 | 71.4 |