Reproduce “Reliability-Aware Sparse Route Memory for Round-Trip Vision-Language Navigation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Bojun Long, Lingfan Bao, Tianhu Peng, Jingcheng Sun, Chengxu Zhou
The paper tests whether a robot can use its outbound route memory to return home during continuous vision-language navigation. It records LiDAR route anchors, then uses them for directional hints, action arbitration, and stopping checks. On simulated M50 routes, conditional Return success rises from 22.0% to 55.1% online; results are limited to one policy and setup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.