ResearchDatasetsStandard Downstream BenchmarksStandard Downstream BenchmarksFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsNatural Language Understanding and ReasoningStandard Downstream Benchmarks Two-Shot (val)56.86ARC-E Accuracy (Normalized)11
Natural Language Understanding and ReasoningStandard Downstream Benchmarks Two-Shot (val)56.86ARC-E Accuracy (Normalized)11