Online Reinforcement Learning on Meta-World
1.343RewardLaGO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| LaGOTask=Easy2026.06 | 1.343 | 8.2 | |
| LaGOTask=Real2026.06 | 1.167 | 23.3 | |
| LaGOTask=Overall Average2026.06 | 1.161 | 15.2 | |
| LaGOTask=Rephrase2026.06 | 1.13 | 23.5 | |
| Vanilla PPOTask=Easy2026.06 | 1.108 | 5.7 | |
| LaGOTask=Hard2026.06 | 1.002 | 5.8 | |
| Vanilla PPOTask=Overall Average2026.06 | 0.84 | 2.7 | |
| Vanilla PPOTask=Hard2026.06 | 0.768 | 0.5 | |
| Vanilla PPOTask=Real2026.06 | 0.741 | 2.3 | |
| Vanilla PPOTask=Rephrase2026.06 | 0.741 | 2.3 |