ResearchTasksAgreement with outcome human labelsFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedCUAVerifierBench Browserbase OM2W (n=106)Universal Verifier (UV)88Accuracy8Apr 9, 2026CUAVerifierBench Internal DatasetUniversal Verifier (UV)81Accuracy8Apr 9, 2026