Reinforcement Learning Control on ManiSkill stack-cube 2
170ReturnClean
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Clean2026.06 | 170 | 4.4 | — | — | — | |
| SWAAPalpha=0.9, rp=0.12026.06 | 141 | — | 7.7 | 4.2 | 4.1 | |
| SWAAP (Random)alpha=0.9, rp=0.12026.06 | 137 | — | 8.8 | 4.2 | — | |
| Direct Model Poisoning2026.06 | 106 | — | 11.6 | — | — |