ResearchDatasetsEditReward-BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsReward ModelingEditReward-Bench85.4PF25Multi-way preference rankingEDITREWARD-BENCH66.2Preference Score (K=2)23Image editing preference evaluationEditReward-Bench63.27Accuracy14Visual Consistency AssessmentEditReward-Bench 2025 (test)80.88Subject Addition Accuracy6