With the continuous development of computer hardware equipment and deep learning technology, it is easier for people to swap faces in videos by currently-emerging multimedia tampering tools, such as ...

Vision-and-Language Navigation (VLN) requires an agent to follow given instructions to navigate. Despite the significant progress, the model trained on seen environments has a performance drop on ...