cs.RONov 6, 2026

BATCH: Behavior Alignment for Cross-Embodiment Learning from Humans

Authors: Zeyi Liu, Shuran Song

Organizations: Stanford University

Abstract

Human demonstrations provide a scalable source of diverse manipulation data, but leveraging them for robot learning remains challenging due to differences in embodiment, viewpoint, action spaces, and temporal structure. We introduce BATCH, a cross-embodiment learning framework that explicitly aligns human and robot behaviors. BATCH learns self-supervised temporal representations with a cycle-consistency objective to establish cross-embodiment correspondences without manual annotation. We further introduce two complementary training objectives that leverage these correspondences within an autoregressive videoaction diffusion model: behavior-aligned co-training and human-to-robot style transfer. On real-world manipulation tasks, BATCH improves generalization to unseen task configurations compared with training on unaligned data and prior cross-embodiment baselines, with additional gains from the style-transfer objective. Our results demonstrate that explicit behavior alignment enables more effective transfer from human demonstrations to robot policies.

Figures & tables

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Cross-Embodiment Transfer via Behavior-Aligned Representations

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

  2. EgoX: Egocentric Cross-Embodiment Manipulation with Embodiment Dreaming

    Nov 12, 2026Yaru Niu, Binghong Chen, Changwei Yao +19Cross-Embodiment Robot LearningCross-Embodiment Transfer

  3. UniLatent: Cross-Embodiment Transfer via Latent Observation Alignment

    Nov 6, 2026Yufei Ding, Yi Du, Weiduo Yuan +5Cross-Embodiment TransferLatent Space Alignment