ResearchTasksRubric satisfaction evaluationRubric satisfaction evaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMLGPT-5-Thinking36.7Claude-4 Sonnet Score24Feb 26, 2026MedicalGPT-5-Thinking50.9Claude-4 Sonnet Score21Feb 26, 2026ArXivGPT-5-Thinking38.1Claude-4 Sonnet Score21Feb 26, 2026