ResearchDatasetsUnified Abstention Benchmark SuiteUnified Abstention Benchmark SuiteFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsQuestion Answering with AbstentionUnified Abstention Benchmark Suite (MMLU, GSM-MC, UMWP, Knowledge Crosswords, HellaSwag, Quail, Misconceptions, Propaganda, BBQ)91.5MMLU Accuracy24
Question Answering with AbstentionUnified Abstention Benchmark Suite (MMLU, GSM-MC, UMWP, Knowledge Crosswords, HellaSwag, Quail, Misconceptions, Propaganda, BBQ)91.5MMLU Accuracy24