Knowledge-based Visual Question Answering on A-OKVQA (val)
68.9AccuracyBLIP VQAv2 + SelTDA
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| BLIP VQAv2 + SelTDAFine-tuned on=VQAv2, Augmentation=SelTDA2023.06 | 68.9 | 1.1 | 8.6 | |
| BLIP VQAv2Fine-tuned on=VQAv22023.06 | 67.8 | — | — | |
| BLIP + SelTDAAugmentation=SelTDA2023.06 | 62.1 | 5 | 1.8 | |
| GPV-22023.06 | 60.3 | — | — | |
| BLIP2023.06 | 57.1 | — | — | |
| KRISP2023.06 | 51.9 | — | — | |
| LXMERT2023.06 | 51.4 | — | — | |
| ViLBERT2023.06 | 49.1 | — | — |