cs.RONov 19, 2026

One View, One Action: Spatially Generalizable VLAs through Observation-Aligned Experts and Cross Graph Consistency

Authors: Jinxi Liu, Ruifeng Zhai, Hao Liu, Guangrun Wang

Organizations: Sun Yat-sen University · X-Era AI Lab

Abstract

Multi-view vision-language-action (VLA) models still face challenges in spatial generalization, particularly under changes in camera viewpoints and robot initial states. These models typically receive action supervision only in the robot base frame, requiring them to implicitly learn the spatial correspondence between observations from each view and base-frame actions. We propose One View, One Action (OVOA-VLA), which learns to predict actions in each view's camera frame, allowing the same physical motion to provide diverse, observation-aligned supervision across views. OVOA-VLA jointly trains a base-frame expert and view-specific action experts, whose robot-state inputs and action targets are expressed in their respective camera frames. Two complementary directed attention graphs model both directions of interaction between the base and view-specific experts, jointly leveraging view-specific action supervision to improve base-frame prediction. Cross Graph Consistency further encourages agreement between the two graphs' base-frame predictions for the same sample. With the same multi-view inputs, the model supports both full and base-only inference. Experiments across multiple simulation benchmarks demonstrate improved manipulation performance and robustness to camera-viewpoint and robot-initialization perturbations. Full and base-only inference achieve average success rates of 88.1% and 85.7% on LIBERO-Plus, respectively. In real-world evaluations, base-only inference raises the average success rates from π0.5π_{0.5}'s 28.3% to 58.3% across three tasks under spatial distribution shifts.

Figures & tables

Appendix figures & tables14 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. OC-VLA++: Monocular Geometry-Guided Cross-View Consistency for Viewpoint-Robust Robotic Manipulation

    Aug 2, 2026Tianyi Zhang, Ziyang Gong, Zhenjie Yang +2Multi-View ConsistencyGeneralization in Robotic Manipulation

  2. The Moving Eye: Enhancing VLA Spatial Generalization via Hybrid Dynamic Data Collection

    Jul 2, 2026Jincheng Tang, Yilong Zhu, Zhengyuan Xie +2Generalization in Robotic ManipulationRobotic Manipulation

  3. UniviewVLA: A Unified Multiview Vision-Language-Action Model with World Modeling

    Jun 19, 2026Tao Xu, Runhao Zhang, Zhijian Huang +7World Models for RoboticsMulti-View Learning