UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling

Boyu Chen, Yi Chen, Lu Qiu, Jerry Bai, Yuying Ge, and Yixiao Ge.

Paper (PDF) · Author-recommended BibTeX · arXiv:2604.19734 · DOI

Abstract

Scaling humanoid foundation models is bottlenecked by the scarcity of robotic data. While massive egocentric human data offers a scalable alternative, bridging the cross-embodiment chasm remains a fundamental challenge due to kinematic mismatches. We introduce UniT (Unified Latent Action Tokenizer via Visual Anchoring), a framework that establishes a unified physical language for human-to-humanoid transfer.

Grounded in the philosophy that heterogeneous kinematics share universal visual consequences, UniT employs a tri-branch cross-reconstruction mechanism: actions predict vision to anchor kinematics to physical outcomes, while vision reconstructs actions to filter out irrelevant visual confounders. Concurrently, a fusion branch synergies these purified modalities into a shared discrete latent space of embodiment-agnostic physical intents.

We validate UniT across two paradigms: 1) Policy Learning (VLA-UniT): By predicting these unified tokens, it effectively leverages diverse human data to achieve state-of-the-art data efficiency and robust out-of-distribution (OOD) generalization on both humanoid simulation benchmark and real-world deployments, notably demonstrating zero-shot task transfer.

2) World Modeling (WM-UniT): By aligning cross-embodiment dynamics via unified tokens as conditions, it realizes direct human-to-humanoid action transfer. This alignment ensures that human data seamlessly translates into enhanced action controllability for humanoid video generation.

Ultimately, by inducing a highly aligned cross-embodiment representation (empirically verified by t-SNE visualizations revealing the convergence of human and humanoid features into a shared manifold), UniT offers a scalable path to distill vast human knowledge into general-purpose humanoid capabilities.

Citation

Use the complete entry below or download the author-recommended BibTeX. The arXiv report number is included in the journal field so that it remains visible with common BibTeX styles.

@article{chen2026unit,
  title   = {{UniT}: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling},
  author  = {Chen, Boyu and Chen, Yi and Qiu, Lu and Bai, Jerry and Ge, Yuying and Ge, Yixiao},
  journal = {arXiv preprint arXiv:2604.19734},
  year    = {2026},
  eprint  = {2604.19734},
  archivePrefix = {arXiv},
  primaryClass  = {cs.RO},
  doi     = {10.48550/arXiv.2604.19734},
  url     = {https://arxiv.org/abs/2604.19734}
}

Recommended citation: Boyu Chen, Yi Chen, Lu Qiu, Jerry Bai, Yuying Ge, and Yixiao Ge. "UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling." arXiv preprint arXiv:2604.19734, 2026.
Download Paper | Download Bibtex