Goal-Conditioned Reinforcement Learning on OGBench pointmaze-medium-stitch v0
100Success RateCD
Evaluation Results
| Method | Links | |
|---|---|---|
| CDMode=Planning Method2025.10 | 100 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 93.4 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 85.4 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 81 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 80.5 | |
| QRLMode=Base Learner2025.10 | 76 | |
| HIQLMode=Base Learner2025.10 | 73 | |
| SAWMode=Base Learner2025.10 | 68.2 | |
| OTAMode=Base Learner2025.10 | 60.6 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 44 | |
| GCIQLMode=Base Learner2025.10 | 18.2 |