ResearchDatasetsAgent Security ScenariosFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsTool-invocation hijacking to RCEAgent Security Scenarios Old LLM Lineup100RCE Success Rate20Tool-invocation hijacking to RCEAgent Security Scenarios New LLM Lineup100Gemini-3.1-pro RCE Success Rate5
Tool-invocation hijacking to RCEAgent Security Scenarios New LLM Lineup100Gemini-3.1-pro RCE Success Rate5