Reproduce “Discrete Model-Based Diffusion for Combinatorial Optimization in Multi-Robot Task Allocation” 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
Haejoon Lee, Thomas Marshall Vielmetti, Xinyi Wang, Dimitra Panagou
The authors introduce Discrete MBD, a training-free diffusion-based method for multi-robot task allocation that searches directly over discrete plans. It uses known plan costs in a Monte Carlo approximation to steer categorical sampling, avoiding an offline-trained denoiser. In a 50-robot, 50-task setting, it runs faster than the reinforcement-learning baseline, but has slightly higher mean makespan.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.