LightNav-0: Eliciting VLM Spatial Intelligence for Generalist Embodied Navigation
cs.RO, cs.AI
Submitted: 2026-08-31
Updated: 2026-09-09
Comments: Technical report
Project page: https://internrobotics.github.io/internvla-n1.github.io
License: http://creativecommons.org/licenses/by/4.0/
The gist: Embodied navigation requires agents to translate heterogeneous goals and visual observations into actions across tasks, environments, and robot embodiments.
Terminology
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.
Sources
- 1st Place Solutions for RxR-Habitat Vision-and-Language Navigation Competition (CVPR 2022)
- On Evaluation of Embodied Navigation Agents
- Qwen3-VL Technical Report
- ObjectNav Revisited: On Evaluation of Embodied Agents Navigating to Objects
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- Robostral Navigate
- Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
- Affordances-Oriented Planning using Foundation Models for Continuous Vision-Language Navigation
- $\pi_\texttt{RL}$: Online RL Fine-tuning for Flow-based Vision-Language-Action Models
- VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning
- PointArena: Probing Multimodal Grounding Through Language-Guided Pointing
- ABot-N0: Technical Report on the VLA Foundation Model for Versatile Embodied Navigation
- Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding
- EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Models
- MolmoAct2: Action Reasoning Models for Real-world Deployment
- Video-R1: Reinforcing Video Reasoning in MLLMs
- OctoNav: Towards Generalist Embodied Navigation
- Gemini Robotics: Bringing AI into the Physical World
- ABot-N1: Toward a General Visual Language Navigation Foundation Model
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- ARC: A Reasoning Recipe for Robot Foundation Models
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies