ResearchBenchmarksWalker-run on DeepMind Control Suite (DMC) (Total Average Return)Walker-run on DeepMind Control Suite (DMC) (Total Average Return)Follow818Average ReturnBRO14.08222.79431.5640.21Jun 29, 2026Evaluation resultsMethodMethodLinksAverage ReturnBRO2026.06818Dual-Flow RL2026.06809TD32026.06794SAC2026.06785BRO-Fast2026.06763QSM2026.06730FlowRL2026.06645D4PG2026.06636QVPO2026.06612DDPG2026.06466MPO2026.06392TRPO2026.0669A2C2026.0645