Vision-Language Navigation on R8R unseen v1.0 (val)
14.2PLRCM(GOAL)+
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| RCM(GOAL)+pre-trained with data augmentation=true2020.05 | 14.2 | 9.6 | 20.3 | 14.9 | 27.7 | 21.9 | 6.4 | |
| RCM(FIDELITY)+pre-trained with data augmentation=true2020.05 | 29.9 | 9.7 | 22.4 | 8.1 | 38.9 | 22.2 | 6.8 | |
| BABYWALK +pre-trained with data augmentation=true2020.05 | 36.8 | 10 | 21 | 6.8 | 46.3 | 29.3 | 9.9 | |
| BABY WALK2020.05 | 38.3 | 10.2 | 20.8 | 6.6 | 45.9 | 28.4 | 9.6 | |
| SEQ2SEQ2020.05 | 43 | 9.9 | 20.1 | 11.2 | 20.6 | 16.3 | 5.6 | |
| SF+pre-trained with data augmentation=true2020.05 | 52.8 | 9.9 | 20.3 | 9.4 | 18.3 | 15.2 | 5 |