Search papers, labs, and topics across Lattice.
This paper introduces MA-VLA, a novel framework for multi-arm collaboration in embodied manipulation that enhances vision-language-action (VLA) models by explicitly assigning atomic actions to individual arms. By implementing a mechanism called Arm Shuffle, which permutes observations and prompts during training, MA-VLA enables role-agnostic instruction following and facilitates compositional generalization to unseen collaboration patterns. Experimental results demonstrate that MA-VLA outperforms existing state-of-the-art VLA models in both simulation and real-world scenarios, highlighting the effectiveness of structured action assignment in improving generalization capabilities.
MA-VLA achieves robust multi-arm compositional generalization, outperforming existing models by effectively enabling role-agnostic behavior in unseen collaboration scenarios.
Multi-arm collaboration is becoming a core capability in embodied manipulation. Recent vision-language-action (VLA) models integrate perception, language, and control, but most represent language as a single global instruction and do not provide an explicit mechanism for assigning and composing arm-specific behaviors. This design limits transfer to collaboration patterns that differ from those observed during training. We present MA-VLA, a unified framework for multi-arm collaboration via atomic action assignment. MA-VLA decomposes cooperative behavior into mid-level atomic prompts and allocates them to individual arms, enabling explicit subgoal specification and compositional reuse across tasks. To reduce reliance on fixed execution roles, we introduce Arm Shuffle, a training-time permutation of the observation, state, and assigned atomic prompts for each arm. This permutation enforces role-agnostic instruction following and supports recomposition into unseen coordination patterns, which we term multi-arm compositional generalization. We also construct a benchmark in which test-time collaboration patterns are absent in training set. Across simulation and real-world evaluations, prior state-of-the-art VLAs largely fail under these unseen collaborations, while MA-VLA consistently succeeds. These results indicate that structured, per-arm atomic action assignment offers a practical route to scalable generalization in multi-arm embodied systems. Code, models, and data are available at https://github.com/zhangzaibin/future-robots
