Qwen-RobotManip技术报告：对齐解锁机器人操作基础模型的规模

摘要

语言和多模态领域的基座模型通过将异构数据统一到相同的范式下并进行规模化的训练，实现了强大的泛化能力。在本报告中，我们探讨了这一规模化策略是否能够应用于机器人操作领域，以实现真正的泛化。这极具挑战性，因为与文本不同，操作数据本质上是异构的、收集成本高昂且多样性有限，这使得数据对齐和规模化训练难以同时实现。我们提出了Qwen-RobotManip——一个基于Qwen-VL构建的可泛化视觉-语言-动作基座模型。Qwen-RobotManip在操作的表征、运动和动作维度上引入了一套统一的对齐框架，使得大规模多源训练变得协调一致而非相互冲突。这种对齐能力反过来又使Qwen-RobotManip能够吸收此前训练范式无法支撑的大规模操作数据。我们构建了一个人机合成流水线，将第一人称手部演示转换为横跨15个平台的机器人轨迹，并设计了一套严格的筛选流水线来调和异构数据集。仅使用开源数据集和人类视频，无需专有数据采集，Qwen-RobotManip构建了约38,100小时的预训练语料库，并展现出涌现式泛化能力，包括零样本指令跟随、对抗扰动的鲁棒性、主动错误恢复以及跨本体迁移。我们发现，现有标准基准难以衡量预训练质量，因此采用了包括RoboCasa365、LIBERO-Plus、EBench、RoboTwin-Clean2Rand、RoboTwin-IF和RoboTwin-XE在内的分布外设定进行评估。Qwen-RobotManip在所有分布外设定下均显著优于包括π0.5在内的先前最佳模型，在RoboChallenge中排名第一且相对性能提升20%，并在包括AgileX ALOHA、Franka、UR和ARX在内的真实机器人平台上得到验证。

English

Foundation models in language and multimodality achieve strong generalization by aligning heterogeneous data under a unified formulation and training at scale. In this report, we investigate whether this scaling recipe can be applied to robotic manipulation to achieve genuine generalization. This is challenging because, unlike text, manipulation data is heterogeneous by nature, expensive to collect, and narrow in diversity, making alignment and scale simultaneously difficult. We present Qwen-RobotManip, a generalizable Vision-Language-Action foundation model built on Qwen-VL. Qwen-RobotManip introduces a unified alignment framework across the representation, motion, and behavioral dimensions of manipulation, making large-scale multi-source training coherent rather than conflicting. This alignment capability in turn enables Qwen-RobotManip to absorb manipulation data at a scale that prior training regimes could not sustain. A human-to-robot synthesis pipeline converts egocentric hand demonstrations into robot trajectories across 15 platforms, and a rigorous curation pipeline harmonizes heterogeneous datasets. Using only open-source datasets and human videos without proprietary data collection, Qwen-RobotManip constructs a ~38,100-hour pretraining corpus and exhibits emergent generalization capabilities, including zero-shot instruction following, robustness to perturbations, reactive error recovery, and cross-embodiment transfer. We find that standard benchmarks fail to capture pretraining quality and instead adopt OOD settings including RoboCasa365, LIBERO-Plus, EBench, RoboTwin-Clean2Rand, RoboTwin-IF, and RoboTwin-XE. Qwen-RobotManip substantially outperforms prior state-of-the-art models, including π0.5, across all OOD settings, ranks 1st in RoboChallenge with a 20% relative improvement, and is validated on real-robot platforms including AgileX ALOHA, Franka, UR, and ARX.