ResearchBenchmarksContinuous Control on SafetyAntVelocity v1Follow3,223.58ReturnPPO-Lag3,002.69443,060.03973,117.3853,174.7303May 12, 2026Evaluation ResultsMethodMethodLinksReturnCVPPO-LagAlgorithm=PPO-LagAlgorithm=PPO-Lag2026.053,223.58-4.96SPMA-ALMAlgorithm=SPMA-ALMAlgorithm=SPMA-ALM2026.053,202.29-2.68PPO-ALMAlgorithm=PPO-ALMAlgorithm=PPO-ALM2026.053,042.762.4CPOAlgorithm=CPOAlgorithm=CPO2026.053,011.19-11.26