ResearchBenchmarksH1balance_simple on Humanoid-BenchFollow354Total ReturnDual-Flow RL-3.7689.12182274.88Jun 29, 2026Evaluation ResultsMethodMethodLinksTotal ReturnDual-Flow RL2026.06354FlowRL2026.06314SAC2026.06231TD32026.06225QVPO2026.06221QSM2026.0654BRO2026.0620BRO-Fast2026.0616DDPG2026.0613A2C2026.0613D4PG2026.0612MPO2026.0612TRPO2026.0610