aiqbench analysis
This research introduces a dual-level structural decomposition framework for bimanual robotic manipulation, addressing the challenge of varying visual relevance and arm coordination modes. By integrating a view-selective visual router with an interaction-aware action mixture-of-experts, the model significantly outperforms monolithic baselines in both simulation and real-world long-horizon tasks. The work demonstrates that explicitly adapting visual conditioning and action generation structures provides a robust inductive bias for complex dual-arm systems.
See Selectively, Act Adaptively: Dual-Level Structural Decomposition for Bimanual Robot Manipulation
arXiv:2606.13279v2 Announce Type: replace Abstract: In bimanual robotic manipulation, task-relevant visual information varies with the task stage and context, while the interaction of the two arms shifts between independent and coordinated modes, making policy learning challenging. However, existing monolithic VLA policies typically process multi-view observations and distinct bimanual interaction patterns without explicitly adapting their visual conditioning or action-generation structure. To address this issue, we propose a bimanual manipulation VLA framework based on Dual-Level Structural Decomposition. The View-Selective Visual Router dynamically adjusts wrist-view contributions to emphasize relevant visual cues, while the Interaction-Aware Action Mixture-of-Experts decomposes action generation into coordinated and arm-wise pathways to adapt to varying bimanual interaction modes. We evaluate the proposed method on six simulated bimanual manipulation tasks in RoboTwin 2.0 and three long-horizon real-world tasks. Our model improves the overall average success rate over a monolithic baseline by 27.7% in simulation and 43.3% in real-world evaluation, while consistently outperforming single-module variants across both settings. These results demonstrate that jointly considering selective visual processing and explicit decomposition of bimanual interaction structures provides an effective inductive bias for robust bimanual manipulation.
Abstract:In bimanual robotic manipulation, task-relevant visual information varies with the task stage and context, while the interaction of the two arms shifts between independent and coordinated modes, making policy learning challenging. However, existing monolithic VLA policies typically process multi-view observations and distinct bimanual interaction patterns without explicitly adapting their visual conditioning or action-generation structure. To address this issue, we propose a bimanual manipulation VLA framework based on Dual-Level Structural Decomposition. The View-Selective Visual Router dynamically adjusts wrist-view contributions to emphasize relevant visual cues, while the Interaction-Aware Action Mixture-of-Experts decomposes action generation into coordinated and arm-wise pathways to adapt to varying bimanual interaction modes. We evaluate the proposed method on six simulated bimanual manipulation tasks in RoboTwin 2.0 and three long-horizon real-world tasks. Our model improves the overall average success rate over a monolithic baseline by 27.7% in simulation and 43.3% in real-world evaluation, while consistently outperforming single-module variants across both settings. These results demonstrate that jointly considering selective visual processing and explicit decomposition of bimanual interaction structures provides an effective inductive bias for robust bimanual manipulation.
Robotics (cs.RO)
arXiv:2606.13279 [cs.RO]
(or arXiv:2606.13279v2 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.13279 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 267 words
Related coverage
See Selectively, Act Adaptively: Dual-Level Structural Decomposition for Bimanual Robot Manipulation
arXiv · Jun 12, 2026
News
Quantum Computing Report · Jun 7, 2026
Hierarchical Skill Retrieval for Data-Efficient Adaptation of Vision-Language-Action Models
arXiv · Aug 27, 2026
Our response to Séb Krier on Plan A
LessWrong · Jul 14, 2026