ResearchDatasetsWHEN2TOOLWHEN2TOOLFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSingle-hop Tool CallingWHEN2TOOL single-hop 1.0 (test)94.3Accuracy90Tool CallingWHEN2TOOL Overall-1Δ Accuracy7Tool CallingWHEN2TOOL Hard-0.8Delta Accuracy (ΔAcc)7Tool CallingWHEN2TOOL Medium-0.7Delta Accuracy7Tool CallingWHEN2TOOL Easy-0.3ΔAcc7