ResearchTasksMulti-task Model EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedAIME24, AIME25, HMMT25, HE+, MBPP+ AverageTeacher Model58.4Average Accuracy16Jun 23, 2026