Vision-and-Language Navigation on Map2seq (test)
52.44TCFLAME
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FLAMEvisual context setting=restricted2024.08 | 52.44 | 5.91 | 67.72 | |
| VELMA (2024)visual context setting=broader2024.08 | 48.7 | 6.8 | 62.37 | |
| ORAR* (2022)visual context setting=restricted (consistent with FLAME)2024.08 | 48.53 | 7.22 | 61.91 | |
| ORAR (2022)visual context setting=broader2024.08 | 47.75 | 6.53 | 62.1 | |
| Loc4Plan (2024)visual context setting=broader2024.08 | 45.3 | 7.2 | — | |
| GA (2019)visual context setting=broader2024.08 | 17 | — | 30.1 | |
| VLN-Trans (2021)visual context setting=broader2024.08 | 17 | — | 29.5 | |
| RCONCAT (2019)visual context setting=broader2024.08 | 14.7 | — | 27.7 | |
| RCONCAT* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 7.63 | 41.22 | 12.03 | |
| GA* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 6.07 | 41.25 | 10.88 |