Goal-Conditioned Reinforcement Learning on OGBench antmaze-large-stitch v0
96.3Success RateGAS
Evaluation Results
| Method | Links | |
|---|---|---|
| GASMode=Planning Method2025.10 | 96.3 | |
| OTA +TTGSMode=TTGS Augmented2025.10 | 91.4 | |
| HIQL +TTGSMode=TTGS Augmented2025.10 | 90.8 | |
| SAW +TTGSMode=TTGS Augmented2025.10 | 86.2 | |
| CDMode=Planning Method2025.10 | 86 | |
| OTAMode=Base Learner2025.10 | 85.4 | |
| HIQLMode=Base Learner2025.10 | 73 | |
| QRL +TTGSMode=TTGS Augmented2025.10 | 43.5 | |
| GCIQL +TTGSMode=TTGS Augmented2025.10 | 30.6 | |
| QRLMode=Base Learner2025.10 | 21 | |
| GCIQLMode=Base Learner2025.10 | 7 | |
| SAWMode=Base Learner2025.10 | 3.1 |