Reinforcement Learning Control on ManiSkill2 lift-cube
193ReturnClean
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Clean2026.06 | 193 | 0.019 | — | — | — | |
| SWAAP (Random)alpha=0.99, rp=0.12026.06 | 178 | — | 0.071 | 0.031 | — | |
| SWAAPalpha=0.99, rp=0.12026.06 | 175 | — | 0.068 | 0.032 | 0.031 | |
| Direct Model Poisoning2026.06 | 136 | — | 0.21 | — | — |