ResearchTasksSafety classification disagreementFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedWildChat ContentGemini 3.10.4Disagreement Rate (per 1k Conversations)30May 26, 2026WildChat IntentGemini 3.10.4Disagreement Rate (per 1k conv)30May 26, 2026