ShahadMAlshalawi/VQAv2-VinVL-BiT-Captions

这是一个包含图像和对应问题的数据集，问题以英文和阿拉伯文两种语言提供。每个问题都有一个或多个答案，答案同样提供英文和阿拉伯文版本，并伴有信心度和唯一标识符。数据集还包含了验证集，可用于模型评估。

This dataset consists of images and corresponding questions provided in both English and Arabic. Each question has one or more answers, which are also available in English and Arabic, along with a confidence score and a unique identifier. The dataset includes a validation set for model evaluation.