ResearchBenchmarksfinger-turn-hard on DMControlFollow717.5Average ScoreSAC340.292438.221536.15634.079Sep 24, 2025Evaluation ResultsMethodMethodLinksAverage ScoreSACTraining timesteps=1M,...Training timesteps=1M, Seeds=52025.09717.5FQLTraining timesteps=1M,...Training timesteps=1M, Seeds=52025.09588.6MEOWTraining timesteps=1M,...Training timesteps=1M, Seeds=52025.09418.2DDPGTraining timesteps=1M,...Training timesteps=1M, Seeds=52025.09381.1TD3Training timesteps=1M,...Training timesteps=1M, Seeds=52025.09354.8