ResearchDatasetsRubricBenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsPairwise AccuracyRubricBench 1.0 (test)81.8Pairwise Accuracy (Chat)13Reward ModelingRubricBench59.7Instruction Following (IF) Score8