ResearchTasksBinary comparison for commonsense plausibilityBinary comparison for commonsense plausibilityFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedViComTe Material 1.0 (test)Mistral91.27Accuracy5Apr 21, 2026ViComTe Shape 1.0 (test)EVA-CLIP94.33Accuracy5Apr 21, 2026ViComTe Color 1.0 (test)GPT-493.29Accuracy5Apr 21, 2026CoDa 1.0 (test)EVA-CLIP95.39Accuracy5Apr 21, 2026