ResearchTasksJailbreak Success EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedLLMail-Inject Scenario 2H&S93Tool Calls3Feb 26, 2026JailbreakSuccess Claude-verifiedKIMI6,643Malware TP Count3Feb 26, 2026Merged Jailbreak Evaluation Set KIMI-verified 1.0 (test)Homotopy-Inspired Prompt Obfuscation6,756Malware TP Count3Feb 26, 2026
Merged Jailbreak Evaluation Set KIMI-verified 1.0 (test)Homotopy-Inspired Prompt Obfuscation6,756Malware TP Count3Feb 26, 2026