ResearchDatasetsMJBenchMJBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsVideo Generation AssessmentMJBench (test)72Accuracy8Video Preference ModelingMJBench Video (evaluation)67.3Tau7Reward ModelingMJBench71.6Accuracy6