AlpachinoNLP/

QTSplus-Dataset

QTSplus-Dataset is a comprehensive dataset for video understanding that provides training signals for long-video multimodal language models. This dataset focuses on visual question answering with both multiple-choice (VSCQ) and free-form (VQA) formats. The scripts used to generate this dataset are available in the official QTSplus-Dataset GitHub Repository.

	Project Overview




	
	About QTSplus-Dataset

QTSplus-Dataset is designed to improve video… See the full description on the dataset page: https://huggingface.co/datasets/AlpachinoNLP/QTSplus-Dataset.