Reproduce “HomeLoop: A Simulation Framework and Benchmark for Robot-IoT Coordination in Everyday Domestic Assistance and Safety” 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
Ramy ElMallah, Marc Bishara, Hyegang Kim, Yuxi Zhang, Chongyu Zhu, Chi-Guhn Lee
HomeLoop tests how household robots respond to ambiguous alerts in simulation. Agents use robot cameras and connected devices to gather evidence and act, while monitors check outcomes and prohibited actions. Across 20 tasks in five families, Astra scored 61.9% on a family-balanced measure. Models often intervened before collecting evidence, but the results do not establish real-home safety.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.