ResearchTasksFew-shot Language UnderstandingFew-shot Language UnderstandingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedARC-C, MMLU, TruthfulQA, Winogrande, HellaSwag, PIQA, MathQA Few-shotFP1626.79ARC-C Score6Jul 2, 2026Language tasks (SST-2, SST-5, SNLI, MNLI, RTE, TREC) few-shot k=16FO91.8Accuracy (SST-2)5May 4, 2026GPT-3 Evaluation Suite Few-shotGPT-3 (Original)48.1Accuracy3Feb 26, 2026
ARC-C, MMLU, TruthfulQA, Winogrande, HellaSwag, PIQA, MathQA Few-shotFP1626.79ARC-C Score6Jul 2, 2026