Goal-Conditioned Reinforcement Learning on OGBench pointmaze-giant-navigate v0
94.8Success RateSAW +TTGS
Evaluation Results
| Method | Links | |
|---|---|---|
| SAW +TTGSMode=TTGS Augmented2025.10 | 94.8 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 91.9 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 88.1 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 73.6 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 70.9 | |
| SAWMode=Base Learner2025.10 | 68.5 | |
| OTAMode=Base Learner2025.10 | 67.2 | |
| QRLMode=Base Learner2025.10 | 65.3 | |
| HIQLMode=Base Learner2025.10 | 43 | |
| GCIQLMode=Base Learner2025.10 | 0 |