cs.RONov 6, 2026

UniLatent: Cross-Embodiment Transfer via Latent Observation Alignment

Authors: Yufei Ding, Yi Du, Weiduo Yuan, Hao Jiang, Dylan Goetting, Lawrence Yunliang Chen, Daniel Seita, Yue Wang

Organizations: Stanford University · University of Southern California · UC Berkeley

Abstract

Robot datasets are costly to collect and are usually tied to a specific embodiment, making it difficult to reuse demonstrations across robots with different geometry, appearance, and kinematics. We introduce UniLatent, a cross-embodiment transfer framework that reduces the embodiment-induced observation gap in a shared latent space. UniLatent renders paired source-target robot observations in simulation and aligns a source visual encoder to a frozen target encoder. A policy trained on encoded source demonstrations can then run directly on target observations, with no pixel-level image translation. Across three simulation tasks and two target robot arms, UniLatent reaches up to 93% zero-shot success, exceeding the strongest pixel-translation baseline by over 30 points on average. In real-world xArm-to-Franka transfer, fine-tuning with only 10 target demonstrations per task yields 27/30 successful rollouts. Code, videos and data are available at https://selina2023.github.io/unilatent/.

Figures & tables

Appendix figures & tables7 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Latent Policy Steering: An Efficient and Flexible Framework for Cross-Embodiment Transfer

    Sep 18, 2026Yiqi Wang, Mrinal Verghese, Jeff SchneiderCross-Embodiment TransferVisuomotor Policy Learning

  2. Cross-Embodiment Transfer via Behavior-Aligned Representations

    Jul 30, 2026Ajay Sridhar, Jensen Gao, Jonathan Yang +3Representation LearningVision-Language-Action Models