ResearchTasksComparative ReasoningComparative ReasoningFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updateddelta-SNLIAlways Tell Me The Odds88.9Accuracy9Apr 28, 2026COMPS 2-ChoiceBatch Calibration (BC)8.5Δ Accuracy6Feb 26, 2026