[2603.00182] Embedding Morphology into Transformers for Cross-Robot Policy Learning

Computer Science > Robotics arXiv:2603.00182 (cs) [Submitted on 26 Feb 2026] Title:Embedding Morphology into Transformers for Cross-Robot Policy Learning Authors:Kei Suzuki, Jing Liu, Ye Wang, Chiori Hori, Matthew Brand, Diego Romeres, Toshiaki Koike-Akino View a PDF of the paper titled Embedding Morphology into Transformers for Cross-Robot Policy Learning, by Kei Suzuki and 5 other authors View PDF HTML (experimental) Abstract:Cross-robot policy learning -- training a single policy to perform well across multiple embodiments -- remains a central challenge in robot learning. Transformer-based policies, such as vision-language-action (VLA) models, are typically embodiment-agnostic and must infer kinematic structure purely from observations, which can reduce robustness across embodiments and even limit performance within a single embodiment. We propose an embodiment-aware transformer policy that injects morphology via three mechanisms: (1) kinematic tokens that factorize actions across joints and compress time through per-joint temporal chunking; (2) a topology-aware attention bias that encodes kinematic topology as an inductive bias in self-attention, encouraging message passing along kinematic edges; and (3) joint-attribute conditioning that augments topology with per-joint descriptors to capture semantics beyond connectivity. Across a range of embodiments, this structured integration consistently improves performance over a vanilla pi0.5 VLA baseline, indicating improved ...