ResearchTasksDiscriminative tasksFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedBigBench 13 tasks (val)UD+58.7Accuracy17Feb 26, 2026RA-QARAMoEA-QA72Accuracy4Mar 9, 2026