ResearchDatasetsLLaVA-WFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsVisual ChatLLaVA-W Bench70.1Score14Multimodal ReasoningLLaVA-W (test)90.6Accuracy12Out-of-Distribution General Visual Question AnsweringLLaVA-W0.755Score6Vision-to-TextLLAVA-W English111.9Score3