ResearchDatasetsSkill-Seen MethodSkill-Seen MethodFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsInput prompt classificationSkill-Seen Method 70B100Accuracy (%)5Input prompt classificationSkill-Seen Method 8B100Accuracy5Input prompt classificationSkill-Seen Method 4B100Accuracy5