ResearchDatasetsRubricBenchRubricBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsPairwise AccuracyRubricBench 1.0 (test)81.8Pairwise Accuracy (Chat)13Reward ModelingRubricBench59.7Instruction Following (IF) Score8