Goal-Conditioned Reinforcement Learning on OGBench visual-antmaze-giant-navigate v0
67.2Success RateGAS
Evaluation Results
| Method | Links | |
|---|---|---|
| GASMode=Planning Method2025.10 | 67.2 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 19 | |
| OTAMode=Base Learner2025.10 | 10 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 8.4 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 7.6 | |
| HIQLMode=Base Learner2025.10 | 5.2 | |
| SAWMode=Base Learner2025.10 | 4.2 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 3.3 | |
| GCIQLMode=Base Learner2025.10 | 0.5 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 0.2 | |
| QRLMode=Base Learner2025.10 | 0.1 |