ResearchTasksAggregate Reasoning EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedMedQA, GSM8K OverallSFT-Oracle76.9Accuracy28May 26, 2026Multi-dataset Reasoning SuiteHCP-MAD82.46Average Accuracy12Apr 14, 2026