Vision-Language Navigation on TOUCHDOWN (test)
1,668TCARC
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| ARCloss=L2STOP2020.09 | 1,668 | 18.84 | 16.34 | 53.5 | 16.34 | — | — | |
| ARC2020.09 | 1,413 | 19.41 | 13.62 | 48.02 | 13.5 | — | — | |
| GA2020.09 | 965 | 21.46 | 9.21 | 47.34 | 9.15 | — | — | |
| RCONCAT2020.09 | 965 | 21.65 | 9.45 | 44.34 | 9.42 | — | — | |
| FLAMEvisual context setting=restricted2024.08 | 40.2 | — | — | — | — | 9.53 | 54.56 | |
| Random2020.09 | 36 | 26.94 | 0.01 | 4.44 | 0 | — | — | |
| Loc4Plan (2024)visual context setting=broader2024.08 | 32.9 | — | — | — | — | 11.5 | — | |
| VLN-Video (2024)visual context setting=broader2024.08 | 31.7 | — | — | — | — | 11.2 | — | |
| ORAR (2022)visual context setting=broader2024.08 | 29.6 | — | — | — | — | 11.79 | 45.3 | |
| VELMA (2024)visual context setting=broader2024.08 | 27.38 | — | — | — | — | 15.03 | 41.93 | |
| ORAR* (2022)visual context setting=restricted (consistent with FLAME)2024.08 | 21.34 | — | — | — | — | 11.02 | 32.44 | |
| ARC+L2S (2020)visual context setting=broader2024.08 | 16.68 | — | — | — | — | 18.84 | — | |
| VLN-Trans (2021)visual context setting=broader2024.08 | 16.2 | — | — | — | — | 20.8 | 27.8 | |
| GA (2019)visual context setting=broader2024.08 | 11.9 | — | — | — | — | 19 | 24.9 | |
| RCONCAT (2019)visual context setting=broader2024.08 | 11.8 | — | — | — | — | 20.4 | 22.9 | |
| RCONCAT* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 8.34 | — | — | — | — | 21.02 | 15.7 | |
| GA* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 6.63 | — | — | — | — | 21.76 | 12.44 |