Goal-conditioned Reinforcement Learning on OGBench scene play (5 tasks) zero-shot
19Average ReturnHILP
Evaluation Results
| Method | Links | |
|---|---|---|
| HILPevaluation_mode=Zero-shot evaluation2026.02 | 19 | |
| One-Step FBevaluation_mode=Zero-shot evaluation2026.02 | 16 | |
| BYOL-gammaevaluation_mode=Zero-shot evaluation2026.02 | 15 | |
| FBevaluation_mode=Zero-shot evaluation2026.02 | 13 | |
| ICVFevaluation_mode=Zero-shot evaluation2026.02 | 8 | |
| One-Step FBevaluation_mode=Zero-shot evaluation2026.02 | 8 | |
| FBevaluation_mode=Zero-shot evaluation2026.02 | 6 | |
| Laplacianevaluation_mode=Zero-shot evaluation2026.02 | 4 | |
| HILPevaluation_mode=Zero-shot evaluation2026.02 | 4 | |
| BYOL-gammaevaluation_mode=Zero-shot evaluation2026.02 | 3 |