ResearchTasksOff-Policy LearningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedWiki10-31K Synthetic tau=2 (test)IPS-GT0.552P@514Feb 26, 2026Wiki10-31K Synthetic tau=1 (test)UIPS55.72P@514Feb 26, 2026Wiki10-31K Synthetic tau=0.5 (test)UIPS0.561P@514Feb 26, 2026Simulation Blocks 2-5 Cross-block averagesMa-style OPL0.722Average Value7Apr 28, 2026