Vision-and-Language Navigation on Room-to-Room (R2R) (test unseen)
63SRVLN-BERT
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| VLN-BERTPre-Training status=Pre-Training2021.07 | 63 | — | — | 57 | |
| EnvDrop + CLIP-VILPre-Training status=No Pre-Training2021.07 | 59 | — | — | 53 | |
| VLN-BERT + OSCARPre-Training status=Pre-Training2021.07 | 57 | — | — | 53 | |
| RelGraphPre-Training status=No Pre-Training2021.07 | 55 | — | — | 52 | |
| AuxRNPre-Training status=Pre-Training2021.07 | 55 | — | — | 50 | |
| FAST-ShortPre-Training status=No Pre-Training2021.07 | 54 | — | — | 41 | |
| PREVALENTPre-Training status=Pre-Training2021.07 | 54 | — | — | 51 | |
| EnvDropPre-Training status=No Pre-Training2021.07 | 51 | — | — | 47 | |
| PRESSPre-Training status=No Pre-Training2021.07 | 49 | — | — | 45 | |
| RegretfulDecoding=greedy, Data Augmentation=true2019.03 | 48 | 5.69 | 56 | 40 | |
| SMNAPre-Training status=No Pre-Training2021.07 | 48 | — | — | 35 | |
| RegretfulPre-Training status=No Pre-Training2021.07 | 48 | — | — | 40 | |
| ALTRPre-Training status=No Pre-Training2021.07 | 48 | — | — | 45 | |
| RCMDecoding=greedy, Data Augmentation=true2019.03 | 43 | 6.01 | 51 | 35 | |
| Self-MonitoringDecoding=greedy, Data Augmentation=true2019.03 | 43 | 5.99 | 55 | 32 | |
| RCMPre-Training status=No Pre-Training2021.07 | 43 | — | — | 38 | |
| Speaker-FollowerDecoding=greedy, Data Augmentation=true2019.03 | 35 | 6.62 | 44 | 28 | |
| S-FollowerPre-Training status=No Pre-Training2021.07 | 35 | — | — | 28 | |
| CGPre-Training status=No Pre-Training2021.07 | 33 | — | — | 30 | |
| RPADecoding=greedy2019.03 | 25 | 7.53 | 33 | 23 | |
| RPAPre-Training status=No Pre-Training2021.07 | 25 | — | — | 23 | |
| Student-forcingDecoding=greedy2019.03 | 20 | 7.85 | 27 | 18 | |
| R2RPre-Training status=No Pre-Training2021.07 | 20 | — | — | 18 | |
| RandomDecoding=greedy2019.03 | 13 | 9.77 | 18 | 12 |