Reproduce “DexBioLab: Benchmarking Dexterous Manipulation with Dynamic Experimental Phenomena in Realistic Laboratories” 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

Xuanye Wu, Yue Shi, Yongcan Yang, Yiqun Wang, Zhaoqi Wan, Dayu Xia, Yanming Shao, Jiezhang Cao and 3 more

DexBioLab presents an interactive laboratory benchmark with 14 tasks across four robot embodiments, combining simulated tools with scanned real-lab backgrounds and controllable visual effects. Its particle approximations for flames, vapor, and bubbles are not physics-coupled. In a tube-transfer experiment using 10 real and 90 simulated demonstrations, Gaussian backgrounds yielded 66% normalized progress versus 17% with standard backgrounds for this task.

Summary by alphaXiv.

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