📝 Publications

Inside Fe₀: What Cross-Embodiment Data Teaches an Embodied Foundation Model

Yizhuo Li*, Xiaoyu Zhang*, Yuying Ge*†, Teng Wang, Boyu Chen, Yi Chen, Bo Liu, Feng Qiu, Yuguo Gan, Hui Zhou, Yixiao Ge

*Co-first authors †Project lead

Technical Blog, 2026

Built on UniT and scaled to large-scale heterogeneous data, Fe0 combines minimal in-domain teleoperation with open-loop and on-robot evaluation across L1–L5, revealing what cross-embodiment data transfers—and where it fails.

Project Page

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†, Yixiao Ge

*Equal contribution †Corresponding author

Preprint, 2026

UniT establishes a unified physical language — a single tokenizer that enables zero-shot human-to-humanoid transfer for both policy learning and world modeling.

Project Page · Paper · Code