ResearchBenchmarksH1crawl on Humanoid-BenchFollow961Total Average ReturnDual-Flow RL129345561777Jun 29, 2026Evaluation ResultsMethodMethodLinksTotal Average ReturnDual-Flow RL2026.06961FlowRL2026.06884SAC2026.06731TD32026.06607QSM2026.06600QVPO2026.06589BRO2026.06428BRO-Fast2026.06263D4PG2026.06226MPO2026.06212TRPO2026.06198DDPG2026.06177A2C2026.06161