ResearchTasksParaphrasing RobustnessFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedELI5 prompts 32-bit payload Llama3.1-8B (test)MC2Mark50.9Bit Accuracy7May 13, 2026