Reproduce “SemPhys: Semantic-Physical Co-Learning for Humanoid Motion Generation via Differentiable World Models” 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
Zejian Kang, Yeqiang Wang, Zekai Chen, Boyang Shen, Mingyu Ma, Bingqian Wu, Fan Zhou, Jianjie Zhang and 4 more
SemPhys trains its language-conditioned motion generator against motion executed by a fixed humanoid controller. It uses a differentiable world model to predict a local segment, while a frozen text–motion matcher scores the complete action with its ordered context. On BONES-SEED, it reaches 79.62 ± 0.06% executed R@1, versus 60.10 ± 0.29% for MDM; these are simulation benchmark results, not a quantitative real-world performance estimate.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.