Reproduce “Closing the Dynamics Gap through Self-Directed Robotic Interaction” 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
Xilun Zhang, Liam Achenbach, Luis Pabon, Matthew Foutter, Marco Pavone
CUE tackles billiards dynamics by building a simulation, tracking ball motion, and selecting shots to test weak predictions. It fits the model to observed trajectories. In a simulated benchmark across 25 hidden-physics conditions, Opus 5.5 CUE reached 0.91 cm error after 30 interactions, versus CEM at 3.20 cm, a 72% reduction. The authors demonstrate real-table deployment, but report no quantitative real-world comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.