ResearchDatasetsPKU BeavertailsFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsReward ModelingPKU Beavertails591,256Training Samples3Reward ModelingPKU beavertails unperturbed68.75Win Rate2