Embodied Navigation on Room A (in-distribution)
97.7Success Rate (SR)Privileged PPO (Upper Bound)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Privileged PPO (Upper Bound)Params (M)=0.49, Inference time (ms)=5.002026.02 | 97.7 | 0.977 | 2.3 | |
| LCLAParams (M)=13.0, Inference time (ms)=74.242026.02 | 90.4 | 0.903 | 9.6 | |
| PELAParams (M)=1.2, Inference time (ms)=69.062026.02 | 62.5 | 0.616 | 37.5 | |
| LeLaN (General Model)Params (M)=181.9, Inference time (ms)=135.742026.02 | 61.1 | 0.606 | 33.7 | |
| LCBCParams (M)=13.7, Inference time (ms)=75.412026.02 | 60.1 | 0.591 | 39.7 |