ResearchBenchmarksMulti-Agent Reinforcement Learning on SMAC 3m → 10mvs11mMulti-Agent Reinforcement Learning on SMAC 3m → 10mvs11mFollow6.4Convergence Steps (×10^5)GCT-MARL4.61616.65828.740.742Jun 23, 2026Evaluation resultsMethodMethodLinksConvergence Steps (×10^5)Final Win RateGCT-MARL2026.066.493LA-QT2026.0611.575Baseline2026.0612.577MAIL2026.0614.893DANN2026.0636.765CORAL2026.0642.368CycleGAN2026.065166