arXiv：cs.AI· Boyu Chen, Yi Chen, Lu Qiu, Jerry Bai, Yuying Ge, Yixiao Ge·· 10 小时前
UniT：统一人类到人形机器人的策略学习与世界模型物理语言框架
UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling
AI 导读
UniT（Unified Latent Action Tokenizer via Visual Anchoring）通过三分支交叉重建机制，构建面向人类到人形机器人迁移的统一物理语言，将异构运动学映射到共享离散潜在空间。
正文
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.
| Comments: | Project page: this https URL ; Author-recommended BibTeX: this https URL . This version adds an ancillary citation file; the manuscript is unchanged |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2604.19734 [cs.RO] |
| (or arXiv:2604.19734v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2604.19734 arXiv-issued DOI via DataCite |
Submission history
From: Boyu Chen [view email]
[v1]
Tue, 21 Apr 2026 17:57:27 UTC (6,651 KB)
[v2]
Sun, 11 Oct 2026 11:18:17 UTC (6,652 KB)
来源：arXiv：cs.AI · arxiv.org