ResearchTasksScientific Reasoning & QAScientific Reasoning & QAFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedScience & QA Domain Multiple DatasetsDVPO4.04Average Accuracy7Feb 26, 2026HLEReinforce++3.61Accuracy7Feb 26, 2026SampleQADVPO3.31Accuracy7Feb 26, 2026