Vision-and-Language Navigation on Map2seq (dev)
0.5695TCFLAME
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| FLAMEvisual context setting=restricted2024.08 | 0.5695 | 5.95 | 0.7136 | |
| VELMA (2024)visual context setting=broader2024.08 | 0.5275 | 6.78 | 0.6645 | |
| ORAR (2022)visual context setting=broader2024.08 | 0.4988 | 5.87 | 0.627 | |
| ORAR* (2022)visual context setting=restricted (consistent with FLAME)2024.08 | 0.4988 | 6.74 | 0.6296 | |
| Loc4Plan (2024)visual context setting=broader2024.08 | 0.48 | 7 | — | |
| VLN-Trans (2021)visual context setting=broader2024.08 | 0.186 | — | 0.311 | |
| GA (2019)visual context setting=broader2024.08 | 0.182 | — | 0.33 | |
| RCONCAT (2019)visual context setting=broader2024.08 | 0.171 | — | 0.307 | |
| RCONCAT* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 0.0825 | 38.71 | 0.1251 | |
| GA* (2019)visual context setting=restricted (consistent with FLAME)2024.08 | 0.065 | 38.74 | 0.1167 |