ResearchTasksNatural Language Understanding and Mathematical ReasoningNatural Language Understanding and Mathematical ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMMLU, GSM8k, HellaSwag, WinoGrande (test)FP1677.18MMLU Accuracy13Mar 4, 2026