Object-goal navigation on AI2-THOR Offline Data 2 (test)
60.5CF-SRTwo-Stage DRL training framework
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Two-Stage DRL training frameworkBase Navigation Model=TDANet2025.02 | 60.5 | 25.8 | |
| Two-Stage DRL training frameworkBase Navigation Model=MJOLNIR-r2025.02 | 55.8 | 21 | |
| TDANetTraining Strategy=Standard2025.02 | 53.5 | 23.4 | |
| MJOLNIR-rTraining Strategy=Standard2025.02 | 47.2 | 17.9 | |
| Two-Stage DRL training frameworkBase Navigation Model=Baseline2025.02 | 30.1 | 9.5 | |
| BaselineTraining Strategy=Standard2025.02 | 27.1 | 8.4 |