Scaling Vision-and-Language Navigation With Offline RL

arXiv:2403.18454v1 Announce Type: new
Abstract: The study of vision-and-language navigation (VLN) has typically relied on expert trajectories, which may not always be available in real-world situations due to the significant effort required to collect them. On the other hand, existing approaches to training VLN agents that go beyond available expert data involve data augmentations or online exploration which can be tedious and risky. In contrast, it is easy to access large repositories of suboptimal offline trajectories. Inspired by research in …

abstract agents arxiv beyond cs.cv data expert language navigation offline scaling study them training type vision vision-and-language world