RoboMorpher: Scaling Robot Morphology Representations for Cross-Embodiment Action Prediction
Learning shared action representations is central to cross-embodiment manipulation. However, identical joint values can produce different spatial configurations across robots, limiting representations learned from joint values alone. We introduce RoboMorpher, a morphology-conditioned action tokenizer that jointly reconstructs joint angles and link poses to associate control coordinates with their kinematic effects. MorphGen supplies scalable kinematic supervision by generating serial-arm descriptions from survey-derived geometric priors. We pretrain RoboMorpher on 40 million morphology-state pairs and integrate the frozen tokenizer into a vision-language-action policy for discrete prediction and continuous refinement. We also introduce EmbodiBench, a ten-task benchmark with 30 generated URDFs per task. With downstream URDFs excluded from tokenizer pretraining, the policy achieves 56.40% success on RoboTwin 2.0 and 23.27% on EmbodiBench, exceeding continuous joint-space control by 9.73 and 15.34 percentage points. Removing link-pose supervision reduces EmbodiBench success by 17.94 percentage points, supporting morphology-aware geometric supervision for joint-action representation learning.
Authors
- Ximin Lyu (ORCID: https://orcid.org/0000-0002-5204-5628)
- Ping Luo
- Yao Mu
- Zanxin Chen
- Yan Qin
- Yue Chen
- Zixuan Li
- Kaixuan Wang
- Renjing Xu
- Weiliang Deng
- Wenwei Lin
- Tianshuo Yang
- Junting Chen
- Tianxing Chen
- Zhixuan Liang
Publication Details
- Journal
- arXiv (Cornell University)
- Published
- 2026-11-18
- DOI
- https://doi.org/10.48550/arxiv.2611.24816
- Primary Topic
- Robot Manipulation and Learning
- Type
- preprint