LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation
2026-08-31 • Robotics
RoboticsArtificial Intelligence
AI summaryⓘ
The authors created LightNav-0, a navigation model that helps robots move by understanding visual scenes and spatial goals without needing task-specific parts. They used vision-language models to capture spatial reasoning and designed a method that connects these insights to robot actions in different settings. Their training included a huge amount of navigation data and combined various learning techniques to handle diverse tasks like following instructions and tracking objects. LightNav-0 showed strong results in virtual tests and worked well on real robots without extra training. This work shows how compact vision-language models can be used as a general foundation for robot navigation.
Embodied navigationVision-language modelsSpatial reasoningToken interfaceVector quantizationReinforcement learningInstruction followingObject navigationZero-shot generalizationRobot embodiments
Authors
Shaoan Wang, Aocheng Luo, Fei Huang, Jingyi Xu, Xiaoyang Wang, Yueyu Wang, Qianli Ma, Fan Yang, Ran Mei, Jia Wei, Jiangpeng Hu, Xuhao Liu, Hongming Chen, Yuanbin Shao, Yiyang Lin, Ziliang Li, Liang Pan, Xinhang Liu, Yuntao Ma, Tingxiang Fan
Abstract
Embodied navigation requires agents to translate heterogeneous goals and visual observations into actions across tasks, environments, and robot embodiments. Modern vision-language models (VLMs) already encode spatial priors for visual grounding, spatial reasoning, and pointing, but these capabilities are rarely elicited directly for robot control. Existing navigation systems instead rely on task- or embodiment-specific components, fragmenting perception, reasoning, and action while offering limited generalization. Here we present LightNav-0, a compact generalist embodied navigation model that elicits the spatial intelligence of a pretrained VLM and aligns it with navigation, without task-specific prediction heads. LightNav-0 represents diverse navigation tasks through a unified token interface: dual-channel pointing expresses task-, scene-, and embodiment-agnostic spatial intent, while a residual vector-quantized action tokenizer maps this intent to precise, embodiment-specific trajectories. Together with temporally aware visual history compression, ER mid-training, supervised fine-tuning, and reinforcement learning, this formulation supports instruction following, open-vocabulary object navigation, and visual tracking within a single model. The navigation training corpus spans 2K+ scenes and 4K+ hours of embodied navigation data. LightNav-ER, the embodied-reasoning checkpoint used to initialize LightNav-0, attains the highest complete-set average across 8 embodied-reasoning benchmarks, while LightNav-0 achieves state-of-the-art monocular success rates across all 10 public navigation simulation settings. Real-world evaluations further demonstrate zero-shot generalization across robot embodiments, diverse scenes, and static and dynamic targets. These results establish compact VLMs as a unified and transferable backbone for generalist embodied navigation.