Natural Language Visual Reasoning on NLVR2 (val)
83.15AccuracyX-VLMclip
Evaluation Results
| Method | Links | |
|---|---|---|
| X-VLMclip2022.10 | 83.15 | |
| EfficientVLM2022.10 | 81.83 | |
| X-VLMclipperformance_target=98%2022.10 | 81.49 | |
| X-VLMsmall2022.10 | 79.34 | |
| OSCARB2022.10 | 78.07 | |
| OSCARBperformance_target=98%2022.10 | 76.51 | |
| ViLT2022.10 | 75.7 | |
| LXMERT2021.07 | 74.9 | |
| DistilDualEnc2022.10 | 74.16 | |
| LXMERT-S2021.07 | 74.03 | |
| MiniVLM2022.10 | 73.71 | |
| VisualBERT2021.07 | 67.4 |