Embedding Morphology into Transformers for Cross-Robot Policy Learning

• URL: http://arxiv.org/abs/2603.00182v1
• Date: Thu, 26 Feb 2026 21:42:34 GMT
• 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,
• Abstract summary: Cross-robot policy learning remains a central challenge in robot learning.<n>We propose an embodiment-aware transformer policy that injects morphology via three mechanisms.<n>Across a range of embodiments, this structured integration consistently improves performance over a vanilla pi0.5 VLA baseline.
• Score: 24.85486808758998
• License: http://creativecommons.org/licenses/by/4.0/
• 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 robustness both within an embodiment and across embodiments.

Related papers

• Structural Action Transformer for 3D Dexterous Manipulation [80.07649565189035]
  Cross-embodiment skill transfer is a challenge for high-DoF robotic hands.<n>Existing methods, often relying on 2D observations and temporal-centric action representation, struggle to capture 3D spatial relations.<n>This paper proposes a new 3D dexterous manipulation policy that challenges this paradigm by introducing a structural-centric perspective.
  arXiv  Detail & Related papers  (2026-03-04T11:38:12Z)
• MOTIF: Learning Action Motifs for Few-shot Cross-Embodiment Transfer [55.982504915794514]
  Cross-embodiment policies typically rely on shared-private architectures.<n>We introduce MOTIF for efficient few-shot cross-embodiment transfer.<n>We show that MOTIF significantly outperforms strong baselines in few-shot transfer scenarios.
  arXiv  Detail & Related papers  (2026-02-14T13:21:40Z)
• ArtGen: Conditional Generative Modeling of Articulated Objects in Arbitrary Part-Level States [9.721009445297716]
  ArtGen is a conditional diffusion-based framework capable of generating articulated 3D objects with accurate geometry and coherent kinematics.<n>Specifically, ArtGen employs cross-state Monte Carlo sampling to explicitly enforce global kinematic consistency.<n>A compositional 3D-VAE latent prior enhanced with local-global attention effectively captures fine-grained geometry and global part-level relationships.
  arXiv  Detail & Related papers  (2025-12-13T17:00:03Z)
• Rethinking the Role of Dynamic Sparse Training for Scalable Deep Reinforcement Learning [58.533203990515034]
  Scaling neural networks has driven breakthrough advances in machine learning, yet this paradigm fails in deep reinforcement learning (DRL)<n>We show that dynamic sparse training strategies provide module-specific benefits that complement the primary scalability foundation established by architectural improvements.<n>We finally distill these insights into Module-Specific Training (MST), a practical framework that exploits the benefits of architectural improvements and demonstrates substantial scalability gains across diverse RL algorithms without algorithmic modifications.
  arXiv  Detail & Related papers  (2025-10-14T03:03:08Z)
• AnyBody: A Benchmark Suite for Cross-Embodiment Manipulation [59.671764778486995]
  Generalizing control policies to novel embodiments remains a fundamental challenge in enabling scalable and transferable learning in robotics.<n>We introduce a benchmark for learning cross-embodiment manipulation, focusing on two foundational tasks-reach and push-across a diverse range of morphologies.<n>We evaluate the ability of different RL policies to learn from multiple morphologies and to generalize to novel ones.
  arXiv  Detail & Related papers  (2025-05-21T00:21:38Z)
• ReCoM: Realistic Co-Speech Motion Generation with Recurrent Embedded Transformer [58.49950218437718]
  We present ReCoM, an efficient framework for generating high-fidelity and generalizable human body motions synchronized with speech.<n>The core innovation lies in the Recurrent Embedded Transformer (RET), which integrates Dynamic Embedding Regularization (DER) into a Vision Transformer (ViT) core architecture.<n>To enhance model robustness, we incorporate the proposed DER strategy, which equips the model with dual capabilities of noise resistance and cross-domain generalization.
  arXiv  Detail & Related papers  (2025-03-27T16:39:40Z)
• Dita: Scaling Diffusion Transformer for Generalist Vision-Language-Action Policy [73.75271615101754]
  We present Dita, a scalable framework that leverages Transformer architectures to directly denoise continuous action sequences.<n>Dita employs in-context conditioning -- enabling fine-grained alignment between denoised actions and raw visual tokens from historical observations.<n>Dita effectively integrates cross-embodiment datasets across diverse camera perspectives, observation scenes, tasks, and action spaces.
  arXiv  Detail & Related papers  (2025-03-25T15:19:56Z)
• Diffusion Transformer Policy [48.50988753948537]
  We propose a large multi-modal diffusion transformer, dubbed as Diffusion Transformer Policy, to model continuous end-effector actions.<n>By leveraging the scaling capability of transformers, the proposed approach can effectively model continuous end-effector actions across large diverse robot datasets.
  arXiv  Detail & Related papers  (2024-10-21T12:43:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.