ResearchTasksCritiqueFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedVISCO 1.0 (test)Human86.5VISCore26Feb 26, 2026CriticBench (test)Qwen3-32B94.1Math Score5May 15, 2026