ResearchTasksMultilingual Commonsense ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedM-HellaswagCAUSAL-NATIVE79.2Accuracy (zh)21Feb 26, 2026X-CSQAMINDMERGER-SOFT45.5Accuracy (SW)10May 19, 2026