ResearchTasksMisalignment monitoringMisalignment monitoringFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMisalignment Monitoring Dataset (test)LLM Judge (Haiku, ind)97AUROC (Deception)6Jun 24, 2026Misalignment Monitoring Dataset (dev)LLM Judge (Haiku)0.962AUROC (Deception)6Jun 24, 2026