Vision-and-Language Navigation on AerialVLN 1.0 (val unseen)
19.1Navigation Error (m) (Full)Human
Evaluation Results
| Method | Links | ||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Human2026.06 | 19.1 | 17.9 | 20.7 | 81.7 | 85.5 | 76.5 | 84.7 | 89.4 | 78.4 | 95.5 | 95.3 | 95.8 | 69.9 | 75.2 | 62.7 | 92.2 | 93.6 | 90.3 | |
| AERIS2026.06 | 79.4 | 72.6 | 88.7 | 25.8 | 29.8 | 20.4 | 44.3 | 51.6 | 34.4 | 64.9 | 74.9 | 51.3 | 15 | 18.5 | 10.2 | 83.1 | 88.4 | 76 | |
| LAG2026.06 | 127.9 | 116.9 | 142.9 | 5.1 | 5.9 | 4 | 10.5 | 12.2 | 8.2 | 27.5 | 31.7 | 21.8 | 2.7 | 3.3 | 1.9 | — | — | — | |
| Random2026.06 | 149.7 | 136.8 | 167.2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | — | — | — | |
| CMA2026.06 | 172.1 | 157.3 | 192.3 | 3.2 | 3.7 | 2.5 | 16 | 18.6 | 12.4 | 34.4 | 39.7 | 27.2 | 1.5 | 1.8 | 1.1 | — | — | — | |
| Seq2Seq2026.06 | 218.9 | 200 | 244.5 | 2.3 | 2.7 | 1.8 | 11.7 | 13.6 | 9.1 | 30.4 | 35.1 | 24 | 1 | 1.2 | 0.7 | — | — | — |