ResearchTasksOpen-form mathematical reasoningOpen-form mathematical reasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedDeepMind MATHEMATICSTRIM42.13Exact-match Accuracy7May 15, 2026SVAMPFull-data Fine-tuning66.12Exact Match Accuracy7May 15, 2026GSM8KS2L52.45Accuracy7May 15, 2026