ResearchBenchmarksMulti-Agent Reinforcement Learning on MPE Push (test)Multi-Agent Reinforcement Learning on MPE Push (test)Follow13.7Final ReturnVDN8.0849.5421112.458Apr 19, 2026Evaluation resultsMethodMethodLinksFinal ReturnVDN2026.0413.7QMIX2026.0412.8IQL2026.0410.9DICG2026.0410.2DCG2026.048.6LLM-Priors2026.048.3