Self-Adaptive Sampling for Efficient Video Question-Answering on Image--Text Models

Benchmark Model Rank Results
visual-question-answering-on-msrvtt-qa-1AIO+MIF#14Accuracy: 0.440
visual-question-answering-on-msrvtt-qa-1AIO+MDF#15Accuracy: 0.438
visual-question-answering-on-msrvtt-qa-1GIT+MDF#16Accuracy: 0.423
visual-question-answering-on-msvd-qa-1GIT+MDF#21Accuracy: 0.469
visual-question-answering-on-msvd-qa-1AIO+MIF#22Accuracy: 0.467