ResearchBenchmarksMeta-Reinforcement Learning on Ant-Dir (ID)Meta-Reinforcement Learning on Ant-Dir (ID)Follow269Average ReturnMetaSTAR67.24119.62172224.38May 30, 2026Evaluation resultsMethodMethodLinksAverage ReturnMetaSTAREpisode=FinalEpisode=Final2026.05269UNICORNEpisode=FinalEpisode=Final2026.05200CSROEpisode=FinalEpisode=Final2026.05143CORROEpisode=FinalEpisode=Final2026.05136FOCALEpisode=FinalEpisode=Final2026.0575