ResearchBenchmarksMulti-agent coordination on Iterated Stag Hunt Reward noiseFollow2.93PoPRATTL1.27641.70572.1352.5643Apr 17, 2026Evaluation ResultsMethodMethodLinksPoPPoARATTLbeta=1beta=12026.042.931.82PPOtype=vanillatype=vanilla2026.042.492.14RATTLbeta=-1beta=-12026.041.343.96