ResearchTasksMulti-disciplinary ReasoningMulti-disciplinary ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedScienceQALLaVA-v1.569.83Accuracy12May 8, 2026HLEGemini-3.037.7Accuracy5Feb 26, 2026