Appendix for Pathdreamer: A World Model for Indoor Navigation

The Val-Seen split contains novel navigation trajectories for environments seen in the training split. As described in the main paper, the Structure Generator and Image Generator models perform significantly better on Val-Seen, since they can memorize the training environments with less generalization required. We observe that generation results maintain high fidelity even at large distances from the location of the input observation. Figure 3 presents cherrypicked examples and Figure 4 presents random examples.