ResearchDatasetsEduFeedbackFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMulti-turn Conversation EvaluationEduFeedback alternate8.3MT-Bench Score20LLM AlignmentEduFeedback152.56TFLOPs20Preference AlignmentEduFeedback-Alternate (test)67.78Pairwise Win Rate (excl. ties)5