ResearchTasksRobust Safety and Utility Evaluation in Federated LearningRobust Safety and Utility Evaluation in Federated LearningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMaliciousGen & WildChatStep-Level81.35Rule Score8Feb 26, 2026MaliciousGen & LMSYS-ChatShadow-Level92.5Rule Compliance8Feb 26, 2026BeaverTails & LMSYS-ChatShadow-Level91.92Rule Score8Feb 26, 2026