ResearchTasksAggregate Reasoning EvaluationAggregate Reasoning EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMedQA, GSM8K OverallSFT-Oracle76.9Accuracy28May 26, 2026Multi-dataset Reasoning SuiteHCP-MAD82.46Average Accuracy12Apr 14, 2026