Goal-Conditioned Reinforcement Learning on OGBench visual-antmaze-large-stitch v0
77.2Success RateGAS
Evaluation Results
| Method | Links | |
|---|---|---|
| GASMode=Planning Method2025.10 | 77.2 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 66.4 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 64.6 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 60.4 | |
| HIQLMode=Base Learner2025.10 | 28.7 | |
| OTAMode=Base Learner2025.10 | 22.1 | |
| SAWMode=Base Learner2025.10 | 6.6 | |
| QRLMode=Base Learner2025.10 | 0.4 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 0.4 | |
| GCIQLMode=Base Learner2025.10 | 0.1 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 0 |