Offline Goal-Conditioned Reinforcement Learning on HumanoidMaze Large-Navigate v0
66Success RateHSVL
Evaluation results
| Method | Links | |
|---|---|---|
| HSVL2026.04 | 66 | |
| CTAdual_analogy=true2026.05 | 60 | |
| HIQL2026.04 | 49 | |
| HIQLdual_analogy=false2026.05 | 49 | |
| HIQLdual_analogy=true2026.05 | 48 | |
| CRL2026.04 | 24 | |
| CRLdual_analogy=false2026.05 | 24 | |
| CRLdual_analogy=true2026.05 | 18 | |
| QRL2026.04 | 5 | |
| GCIVLdual_analogy=false2026.05 | 3 | |
| GCIVLdual_analogy=true2026.05 | 3 | |
| GCIVL2026.04 | 2 | |
| GCIQL2026.04 | 2 | |
| GCBC2026.04 | 1 |