Vision-and-Language Navigation on map2seq unseen (dev)
3.7nDTWRConcat
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| RConcat2022.03 | 3.7 | — | — | 2 | |
| GA2022.03 | 3.9 | — | — | 1.8 | |
| VLN Transformer2022.03 | 6.2 | — | — | 3.6 | |
| ORARImage features=ResNet 4th-to-last, Heading delta=false, Junction type=false2022.03 | 8.9 | — | — | 5 | |
| ORARImage features=ResNet 4th-to-last, Junction type=false2022.03 | 13.1 | — | — | 7.4 | |
| ORARImage features=ResNet 4th-to-last, Heading delta=false2022.03 | 33 | — | — | 22 | |
| ORARImage features=ResNet pre-final2022.03 | 34.6 | — | — | 24.2 | |
| ORARImage features=ResNet 4th-to-last2022.03 | 41 | — | — | 27.6 | |
| GA2022.03 | — | 1.4 | 34.3 | — | |
| ORAR full modelBackbone=ResNet pre-final2022.03 | — | 22.3 | 15.6 | — | |
| ORAR full modelBackbone=ResNet 4th-to-last2022.03 | — | 25.8 | 11.9 | — | |
| ORAR full modelBackbone=ResNet 4th-to-last, Ablation=no heading delta2022.03 | — | 20.4 | 16.8 | — | |
| ORAR full modelBackbone=ResNet 4th-to-last, Ablation=no junction type2022.03 | — | 4.3 | 28.9 | — | |
| RConcat2022.03 | — | 1.2 | 33.1 | — |