Inspired by work in computer vision, we propose a novel data augmentation approach that targets low-frequency words by generating new sentence pairs ...
確定! 回上一頁