Goal-Conditioned Reinforcement Learning on OGBench humanoidmaze-large-stitch v0
80.6Success RateGAS
Evaluation Results
| Method | Links | |
|---|---|---|
| GASMode=Planning Method2025.10 | 80.6 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 75.6 | |
| CDMode=Planning Method2025.10 | 72 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 68.9 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 65.1 | |
| OTAMode=Base Learner2025.10 | 58.2 | |
| HIQLMode=Base Learner2025.10 | 28.6 | |
| SAWMode=Base Learner2025.10 | 11.6 | |
| QRLMode=Base Learner2025.10 | 3.9 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 3.1 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 2 | |
| GCIQLMode=Base Learner2025.10 | 0.7 |