Vision-and-Language Navigation on CVDN (unseen val)
6.16SPLNaviLLM
Evaluation results
| Method | Links | |
|---|---|---|
| NaviLLMModel Training Strategy=Unified Model For All Tasks2023.12 | 6.16 | |
| VLN-PETLModel Training Strategy=Separate Model For Each Task2023.12 | 5.69 | |
| VLN-SIGModel Training Strategy=Separate Model For Each Task2023.12 | 5.52 | |
| HAMTModel Training Strategy=Separate Model For Each Task2023.12 | 5.13 | |
| MT-RCM+EnvModel Training Strategy=Unified Model For All Tasks2023.12 | 4.65 | |
| HOPModel Training Strategy=Separate Model For Each Task2023.12 | 4.41 | |
| PREVALENTModel Training Strategy=Separate Model For Each Task2023.12 | 3.15 |