ResearchBenchmarksMulti-task Reinforcement Learning on DIBS Multi-environment Suite (train)Multi-task Reinforcement Learning on DIBS Multi-environment Suite (train)Follow10.83Relative tasks solved ratio (vs DIBS)MAML2.70764.81636.9259.0337May 30, 2026Evaluation resultsMethodMethodLinksRelative tasks solved ratio (vs DIBS)MAML2026.0510.83BC2026.053.94GenRL2026.053.82VariBAD2026.053.02