Speaker-Follower Models for Vision-and-Language Navigation
2018·Arxiv