ResearchDatasetsTool benchmarkTool benchmarkFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsReasoning Quality Evaluation120-tool benchmark 500 tasks simulated4.43Mean Score5