ResearchDatasetsI-BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsVLM-Action-JudgeI-Bench (test)2.557IF Score7Key-Mouse-FollowingI-Bench (test)20.62Accuracy (Full)7Interactive World ModelingI-Bench 1.0 (test)87.1SC7Interactive Video GenerationI-Bench User Study (test)4.05Action Following6