R2R

The dataset used in the paper for vision-and-language navigation tasks.

BibTex: