Agricultural Vision-and-Language Navigation on A2A high-complexity subtask >= 3
80SRHuman
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| HumanMDE Representation=-2025.12 | 80 | 0.82 | — | — | |
| MDE-AgriVLNMDE Representation=Depth Map2025.12 | 23 | 4.81 | 1,739 | 170 | |
| MDE-AgriVLNMDE Representation=Hybrid2025.12 | 21 | 5.1 | 2,585 | 194 | |
| MDE-AgriVLNMDE Representation=Depth Matrix2025.12 | 19 | 5.17 | 2,257 | 199 | |
| AgriVLNMDE Representation=-2025.12 | 16 | 5.18 | 1,401 | 144 | |
| DILLM-VLNMDE Representation=-2025.12 | 10 | 5.56 | 1,279 | 132 | |
| FixedMDE Representation=-2025.12 | 6 | 6.32 | — | — | |
| RandomMDE Representation=-2025.12 | 4 | 6.74 | — | — | |
| SIA-VLNMDE Representation=-2025.12 | 2 | 6.24 | 1,327 | 177 |