DM0: An Embodied-Native Vision-Language-Action Model towards Physical AI

TL;DR

DM0 introduces a unified vision-language-action framework for Physical AI, integrating diverse data sources and reasoning strategies to improve embodied manipulation and navigation tasks.

Contribution

It presents a comprehensive three-stage training pipeline and a novel embodied spatial scaffolding method for better physical grounding and reasoning.

Findings

Achieves state-of-the-art results on RoboChallenge benchmark.

Effectively combines high-level reasoning with low-level control.

Demonstrates improved generalization in embodied tasks.

Abstract

Moving beyond the traditional paradigm of adapting internet-pretrained models to physical tasks, we present DM0, an Embodied-Native Vision-Language-Action (VLA) framework designed for Physical AI. Unlike approaches that treat physical grounding as a fine-tuning afterthought, DM0 unifies embodied manipulation and navigation by learning from heterogeneous data sources from the onset. Our methodology follows a comprehensive three-stage pipeline: Pretraining, Mid-Training, and Post-Training. First, we conduct large-scale unified pretraining on the Vision-Language Model (VLM) using diverse corpora--seamlessly integrating web text, autonomous driving scenarios, and embodied interaction logs-to jointly acquire semantic knowledge and physical priors. Subsequently, we build a flow-matching action expert atop the VLM. To reconcile high-level reasoning with low-level control, DM0 employs a hybrid training strategy: for embodied data, gradients from the action expert are not backpropagated to the VLM to preserve generalized representations, while the VLM remains trainable on non-embodied data. Furthermore, we introduce an Embodied Spatial Scaffolding strategy to construct spatial Chain-of-Thought (CoT) reasoning, effectively constraining the action solution space. Experiments on the RoboChallenge benchmark demonstrate that DM0 achieves state-of-the-art performance in both Specialist and Generalist settings on Table30.