ResearchTasksLong-generation reasoningLong-generation reasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMATH500ParisKV0.93pass@112Feb 26, 2026GPQA DiamondParisKV72.22pass@112Feb 26, 2026