PrimitiveVLA: Learning Reusable Motion Primitives for Efficient and Generalizable Robotic Manipulation
Authors: Yutai Li, Shaohui Peng, Jiaming Guo, Di Huang, Zihao Zhang, Yuxuan Guo, Yunkai Gao, Siming Lan, +3 more
Organizations: State Key Lab of Processors, Institute of Computing Technology, CAS · Jiangsu Key Laboratory of AI for Industries, Institute of AI for Industries, CAS · University of Chinese Academy of Sciences · Cambricon Technologies · Intelligent Software Research Center, Institute of Software, CAS · University of Science and Technology of China
Vision-Language-Action (VLA) models offer a promising paradigm for generalist robotic policies, yet their adaptation is hindered by data inefficiency and poor generalization. We argue that these bottlenecks stem from the prevailing Direct Instruction-to-Control Mapping, which forces models to memorize monolithic trajectories rather than reusable motion patterns, i.e., primitives. We propose PrimitiveVLA, a framework that shifts this paradigm toward a Primitive-Centric Disassemble & Assemble paradigm. Supported by a shared Multimodal Canonical Representation (MCR), PrimitiveVLA unifies two phases: (1) Fine-tuning-phase Disassembly, which uses an automated pipeline to disassemble demonstrations into reusable primitives; and (2) Inference-phase Assembly, which employs a VLM-based planner and an LLM-generated switch module for robust closed-loop execution. By disassembling tasks into reusable primitives, PrimitiveVLA enables VLA models to learn invariant motion patterns instead of task-specific trajectories. Extensive experiments show that our framework improves data efficiency and achieves superior zero-shot generalization across unseen and long-horizon tasks.