ResearchBenchmarksJoint Passage on VMASFollow-2.73Mean Episodic RewardCoHetteam-116.8596-87.2298-57.6-27.9702Aug 12, 2024Evaluation ResultsMethodMethodLinksMean Episodic RewardCoHetteamEnvironment Steps=2×10^5Environment Steps=2×10^52024.08-2.73CoHetselfEnvironment Steps=2×10^5Environment Steps=2×10^52024.08-9.11HetGPPOEnvironment Steps=2×10^5Environment Steps=2×10^52024.08-55.1IPPOEnvironment Steps=2×10^5Environment Steps=2×10^52024.08-112.47