BnB

该数据集是一个从Airbnb房源收集的大规模、多样化的领域内数据集，包含了图像与标题对以及生成的路径说明对。此外，数据集还包括结构化数据，如设施、总体描述、评论、位置和租赁价格。其中95%的数据用于训练，5%用于验证。该数据集规模宏大，拥有超过713,000个图像与标题对以及300万张图片，旨在支持视觉与语言导航（Vln）任务。

This dataset is a large-scale, diverse in-domain dataset collected from Airbnb listings, comprising image-title pairs and generated path description pairs. Additionally, the dataset includes structured data such as amenities, overall descriptions, reviews, locations, and rental prices. 95% of the data is used for training, while the remaining 5% is reserved for validation. Boasting over 713,000 image-title pairs and 3 million images in total, this dataset is designed to support Vision-and-Language Navigation (VLN) tasks.