ResearchTasksCognition and ReasoningCognition and ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMMMU (val)InternVL3.5-20B-A4B72.6Score28May 12, 2026GQATask Arithmetic0.623Score16Feb 26, 2026OKVQATask Arithmetic61.92Score16Feb 26, 2026