Goal-Conditioned Reinforcement Learning on Humanoidmaze giant-navigate v0
92Success RateOTA
Evaluation results
| Method | Links | |
|---|---|---|
| OTAArchitecture=Hierarchical2026.05 | 92 | |
| LAVLArchitecture=Hierarchical2026.05 | 83 | |
| HIQLArchitecture=Hierarchical2026.05 | 12 | |
| CGCIVLArchitecture=Hierarchical2026.05 | 10 | |
| CRLArchitecture=Non-hierarchical2026.05 | 3 | |
| QRLArchitecture=Non-hierarchical2026.05 | 1 | |
| GCBCArchitecture=Non-hierarchical2026.05 | 0 | |
| GCIVLArchitecture=Non-hierarchical2026.05 | 0 | |
| GCIQLArchitecture=Non-hierarchical2026.05 | 0 |