ResearchBenchmarksDexterous Manipulation on Two-Arms ReorientationFollow35.3Average PerformanceCPO0.05449.204718.35527.5053Mar 2, 2026Evaluation ResultsMethodMethodLinksAverage PerformanceCPOEnvironment steps=2 ×...Environment steps=2 × 10^102026.0335.3PBTEnvironment steps=2 ×...Environment steps=2 × 10^102026.0326.43SAPGEnvironment steps=2 ×...Environment steps=2 × 10^102026.035.11PPOEnvironment steps=2 ×...Environment steps=2 × 10^102026.031.41