ResearchTasksMulti-domain Language ModelingMulti-domain Language ModelingFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedMGSM, HumanEval, and MedMCQA composite (test)LoRAFlow46.1Average Score (Composite)9Feb 26, 2026