Exploring the Necessity of Visual Modality in Multimodal Machine Translation using Authentic Datasets

Open in new window