ResearchDatasetsTool selection benchmarkTool selection benchmarkFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsIntent ShiftingTool selection benchmark80Certified Upper Bound (pu)6Privilege EscalationTool selection benchmark82Certified Upper Bound6Distractor SelectionTool selection benchmark30Certified Upper Bound (pu)6