ResearchTasksSelective VerificationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast Updated5,000-record held-out natural-prevalence verification (Overall)Complete verification layer84.4Selective Accuracy1Jun 1, 20265,000-record held-out natural-prevalence verification (Punjabi)Complete verification layer83.2Selective Accuracy1Jun 1, 2026Natural-Prevalence Verification 5,000-record held-out (Hindi)Complete verification layer84.6Selective Accuracy1Jun 1, 2026natural-prevalence verification 5,000-record held-out (English)Complete verification layer85.4Selective Accuracy1Jun 1, 2026
5,000-record held-out natural-prevalence verification (Overall)Complete verification layer84.4Selective Accuracy1Jun 1, 2026
5,000-record held-out natural-prevalence verification (Punjabi)Complete verification layer83.2Selective Accuracy1Jun 1, 2026
Natural-Prevalence Verification 5,000-record held-out (Hindi)Complete verification layer84.6Selective Accuracy1Jun 1, 2026
natural-prevalence verification 5,000-record held-out (English)Complete verification layer85.4Selective Accuracy1Jun 1, 2026