ResearchTasksSafety Guardrail ClassificationSafety Guardrail ClassificationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedCHILLGuard 1.0 (test)CHILLGuard-4B90.01F1 Score (Category A1)16Jun 16, 2026630-scenario real-world benchmark (independent set)AgentTrust v0.595.4Verdict Accuracy5May 7, 2026