ResearchTasksLinguistic Minimal Pair EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedBLiMP (test)Sonnet100NPI lic. (2)28Feb 26, 2026BLiMP, SLING, and RuBLiMP Combined (test)Llama (GPb+CT)0.928Average Score4Feb 26, 2026