Open-vocabulary Video Question Answering: A New Benchmark for Evaluating the Generalizability of Video Question Answering Models

Benchmark Model Rank Results
visual-question-answering-on-msrvtt-qaFrozenBiLM+#7Accuracy: 0.470
visual-question-answering-on-msrvtt-qaJustAsk+#20Accuracy: 0.418
visual-question-answering-on-msrvtt-qaAll-in-one+#21Accuracy: 0.395
visual-question-answering-on-msvd-qaFrozenBiLM+#8Accuracy: 0.558
visual-question-answering-on-msvd-qaVIOLET+#17Accuracy: 0.495
visual-question-answering-on-msvd-qaJustAsk+#20Accuracy: 0.477
visual-question-answering-on-msvd-qaAll-in-one+#24Accuracy: 0.438
video-question-answering-on-activitynet-qaFrozenBiLM+#14Accuracy: 44.8
video-question-answering-on-activitynet-qaAll-in-one+#20Accuracy: 40.0
video-question-answering-on-activitynet-qaVIOLET+#21Accuracy: 39.7