ResearchTasksMultitask EvaluationFor AI assistantsMarkdownContent indexMultitask EvaluationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedAggregate (MMLU, GSM8K, SQuAD V2, Amazon, 20News)ShadowPEFT w/ pretrained shadow77.11Average Score16Apr 23, 2026
Aggregate (MMLU, GSM8K, SQuAD V2, Amazon, 20News)ShadowPEFT w/ pretrained shadow77.11Average Score16Apr 23, 2026