UCAG-P is introduced, a camera-centric unified action formulation that structurally aligns heterogeneous embodied datasets into a shared geometric action space, and allows a shared VLA policy to learn transferable manipulation geometry while retaining embodiment-specific controllability.
Abstract
Scaling generalist vision-language-action (VLA) policies is severely bottlenecked by the inherent heterogeneity of embodied data, which spans diverse robot morphologies, camera configurations, and low-level action spaces. Existing paradigms typically address this mismatch through explicit action retargeting, human-to-robot video synthesis, or dataset-specific adaptation branches, fundamentally hindering the joint learning of a unified policy. We introduce UCAG-P, a camera-centric unified action formulation that structurally aligns heterogeneous embodied datasets into a shared geometric action space. Rather than treating robot-specific commands as the shared policy target, UCAG-P represents manipulation through camera-observable anchor motion in image and camera-frame coordinates, treating robot arms, humanoids, and human hands as different embodiments of a common action schema. A geometry-conditioned action translator combines predicted motion with target-embodiment kinematics to produce executable controls. The resulting decoupled architecture allows a shared VLA policy to learn transferable manipulation geometry while retaining embodiment-specific controllability. UCAG-P is trained on 4.03K hours of robot and simulation data and 2.34K hours of human demonstrations. A single checkpoint reaches 98.3% on LIBERO, 88.7% and 89.2% on RoboTwin Easy and Hard, 82.0% zero-shot on LIBERO-Plus, and 62.0% on RoboCasa GR-1, without benchmark-specific fine-tuning.
This work introduces Universal Manipulation Representation (UMR), a unified action representation that enables zero-shot skill transfer from human demonstrations to heterogeneous robots and complements it with a Data-Efficient Strategy (DES) that diversifies object configurations through stage-aware point-cloud editing...
Song Liu, Lin-Ying Li, Yan-Shun Zhao et al.· arXiv.org· 0 citations
This work proposes VLAct, a VLA-oriented VLM backbone trained on broad, heterogeneous, multi-embodiment robot data before task-specific fine-tuning, showing that representation-centric continued pre-training can deliver highly competitive performance under a modest compute budget and is an important independent axis of...
Sen-Qiao Yang, Cheng-Yao Wang, Yu-Xin Chen et al.· 8 citations
GALA combines visual latent actions that capture scene-level dynamics with geometric latent actions that capture shared fine-grained end-effector articulation, providing effective supervision for VLA pretraining from multi-embodiment data, including action-free ego-centric human videos.
Yi-Chen Liu, Pu-Zhen Yuan, Xiang-Pei Zhu et al.· 1 citation
This work introduces a shared visual action interface, called action views, which render target joint configurations from four prescribed virtual cameras using URDF-based forward kinematics, and proposes a training-free, URDF-constrained multiview recovery mechanism, without a learned embodiment-specific decoder.
Xiang-Yu Zhu, Jin Xu, Yue (Sophie) Guo et al.· 0 citations
AnyWorld is proposed, a cross-embodiment world modeling framework that expands a single human interaction into diverse robot-native rollouts without paired human-robot demonstrations and enables independent recomposition of embodiment, viewpoint, and scene factors, allowing a single model to generate many robot-domain...
Cheng Chen, J. Bai, Jiacheng Wei et al.· 3 citations
This work presents SkelWAM, a skeleton-guided world-action model that couples perception and control through one explicit geometric representation for single-source cross-embodiment manipulation, and introduces LIBERO-Cross10, a source-only cross-embodiment transfer benchmark covering ten tasks and ten target embodimen...