ResearchTasksResearch Paper Reasoning and ComprehensionResearch Paper Reasoning and ComprehensionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedArxivRollBench 2025a (val)google/gemini-2.5-flash44.3Valid Accuracy38May 26, 2026