ResearchBenchmarksReinforcement Learning on Genesis T1 + G1 + Go1 + Go2Reinforcement Learning on Genesis T1 + G1 + Go1 + Go2Follow0.85IQMABD-NET0.31960.45730.5950.7327Mar 19, 2026Evaluation resultsMethodMethodLinksIQMMedianMeanABD-NETTraining Algorithm=PPOTraining Algorithm=PPO2026.030.850.860.83SWATTraining Algorithm=PPOTraining Algorithm=PPO2026.030.790.830.75GNNTraining Algorithm=PPOTraining Algorithm=PPO2026.030.760.760.74BOTTraining Algorithm=PPOTraining Algorithm=PPO2026.030.690.670.7RODRIGUESTraining Algorithm=PPOTraining Algorithm=PPO2026.030.450.450.45MLPTraining Algorithm=PPOTraining Algorithm=PPO2026.030.340.320.4