ResearchTasksTool-augmented agentic reasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedFull Evaluation Corpus (test)Sponge Tool Attack (STA)2.99Reward8Feb 26, 2026Evaluation Corpus High-BudgetSponge Tool Attack24.23Delta Cap Hits (%)6Feb 26, 2026Evaluation Corpus Low-BudgetSponge Tool Attack26.54Delta Cap Hits (%)6Feb 26, 2026