ResearchBenchmarksH1sit_hard on Humanoid-BenchFollow837Total Average ReturnDual-Flow RL-27.24197.13421.5645.87Jun 29, 2026Evaluation ResultsMethodMethodLinksTotal Average ReturnDual-Flow RL2026.06837FlowRL2026.06728QVPO2026.06696QSM2026.06399BRO-Fast2026.0618BRO2026.0617D4PG2026.0617DDPG2026.0615TD32026.0614MPO2026.0611SAC2026.0610A2C2026.068TRPO2026.066