Ranking Distillation for Open-Ended Video Question Answering with Insufficient Labels

arXiv:2403.14430v1 Announce Type: new
Abstract: This paper focuses on open-ended video question answering, which aims to find the correct answers from a large answer set in response to a video-related question. This is essentially a multi-label classification task, since a question may have multiple answers. However, due to annotation costs, the labels in existing benchmarks are always extremely insufficient, typically one answer per question. As a result, existing works tend to directly treat all the unlabeled answers as negative labels, …

abstract annotation arxiv classification costs cs.cv distillation however labels multiple paper question question answering ranking set type video