ResearchTasksSpeech-Visual Question AnsweringSpeech-Visual Question AnsweringFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedVQA Speech-converted v2Single-Modality-Expert-Task332.95Accuracy12Feb 27, 2026OK-VQA Speech-convertedUni-MoE29.04Accuracy12Feb 27, 2026A-OKVQA Speech-convertedUni-MoE0.2Accuracy12Feb 27, 2026