Visual Navigation On Cooperative Vision And 1
評価指標
dist_to_end_reduction
spl
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | dist_to_end_reduction | spl | Paper Title | Repository |
|---|---|---|---|---|
| PREVALENT | 2.44 | 0.24 | - | - |
| Test2-NDH | 3.44 | 0.09 | - | - |
| n151-7200-trusted | 3.14 | 0.01 | - | - |
| BabyWalk | 3.65 | 0.11 | BabyWalk: Going Farther in Vision-and-Language Navigation by Taking Baby Steps | |
| HESNet | 3.26 | 0.11 | - | - |
| Environment-agnostic Multitask Learning | 3.91 | 0.17 | Environment-agnostic Multitask Learning for Natural Language Grounded Navigation | |
| s-agent (NDH-Full) | 5.27 | 0.15 | - | - |
| Pansy | 1.76 | 0.15 | Vision-and-Dialog Navigation | |
| early to bed | 6.03 | 0.06 | - | - |
| Seq2Seq Baseline | 2.35 | 0.16 | Vision-and-Dialog Navigation | |
| VLN-PETL | 6.13 | 0.07 | VLN-PETL: Parameter-Efficient Transfer Learning for Vision-and-Language Navigation | |
| Best SPL | 2.4 | 0.25 | - | - |
| BabyWalk (r2r-pretrain) | 4.46 | 0.23 | BabyWalk: Going Farther in Vision-and-Language Navigation by Taking Baby Steps | |
| aux | 2.75 | 0.11 | - | - |
| NaviLLM | 7.90 | 0.09 | Towards Learning a Generalist Model for Embodied Navigation | |
| HAMT | 5.58 | 0.07 | - | - |
| ed | 3.3 | 0.14 | - | - |
| VISITRON | 3.08 | 0.22 | - | - |
| SCoA | 3.37 | 0.15 | - | - |
0 of 19 row(s) selected.