ResearchTasksLong-horizon recipe evaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast Updatedbpb Linux 12h (val)greedy_winner_10min_s1596.156Mean Val Bpb6Jun 5, 2026bpb Windows 12h (val)predeclared_control_c100.954Mean bpb6Jun 5, 2026