Reproduce “Learning from All Vehicles” 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

Dian Chen, Philipp Krähenbühl

LAV uses nearby vehicles’ trajectories as additional training examples, addressing gaps in routine ego-vehicle logs. It learns a shared map-like representation, trains a privileged planner on observed paths, then distills that learning into an ego-sensor policy. In CARLA simulation, it ranks first with a 61.85 Driving Score and 94.46 Route Completion, but this does not establish real-world safety.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.