ResearchTasksBenign ComplianceBenign ComplianceFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedWJB (WildJailbreak)Base99.2Compliance Rate15Jul 7, 2026XSTestELS97.6Comply Score12Jun 9, 2026XSTest vani benignLlama3.1-8B-IT-AB98.8Comply Score10Jul 7, 2026XSTest VanillaQwen2.5-14B-IT98.4LC Winrate9May 12, 2026WildJailBreak (WJB) Vanilla BenignSELF-REDTEAM100Compliance Rate9May 12, 2026