Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System

Hugging Face Daily Papers Papers
navigation-model multi-task-training zero-shot-generalization robotics vision-language parameterized-interface

Summary

Qwen-RobotNav is a scalable navigation model with a parameterized interface enabling dynamic task modes and observation parameters, achieving state-of-the-art performance through multi-task training and zero-shot generalization to real-world robotics.

Agentic navigation systems require a base navigation model whose observation strategy can be externally reconfigured at inference time, because instruction following, object search, target tracking, and autonomous driving share the same perception-planning backbone yet demand fundamentally different strategies for consuming the visual stream. We present Qwen-RobotNav, a scalable navigation model built on Qwen-RobotNav that addresses it through a parameterised interface with two complementary dimensions: multiple task modes that select the navigation behaviour, and controllable observation parameters (e.g., token budget, per-camera weights) that govern how visual history is encoded. With training-time randomization over all parameters, Qwen-RobotNav is robust to any inference-time configuration requiring zero architectural modification to the Qwen-RobotNav backbone. We train Qwen-RobotNav on 15.6M samples; co-training with vision-language data prevents the collapse into reactive action-sequence mappers observed in trajectory-only training. The parameterised interface also makes Qwen-RobotNav a natural building block for agentic systems: for long-horizon scenarios, an upper-level planner decomposes goals into sub-tasks and dynamically switches Qwen-RobotNav's task mode and context strategy mid-episode, composing complex behaviours from repeated calls to the same model. Extensive experiments show that Qwen-RobotNav sets new state-of-the-art results across major navigation benchmarks. The model exhibits favourable scaling from 2B to 8B parameters, with joint multi-task training developing a shared spatial-planning substrate that transfers across task families, and demonstrates strong zero-shot generalisation to real-world robots across diverse environments.
Original Article
View Cached Full Text

Cached at: 06/29/26, 10:05 PM

Paper page - Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System

Source: https://huggingface.co/papers/2606.18112 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

Qwen-RobotNav is a scalable navigation model with a parameterized interface enabling dynamic task modes and observation parameters, achieving state-of-the-art performance through multi-task training and zero-shot generalization to real-world robotics.

Agentic navigation systems require a basenavigation modelwhose observation strategy can be externally reconfigured at inference time, because instruction following, object search, target tracking, and autonomous driving share the same perception-planning backbone yet demand fundamentally different strategies for consuming the visual stream. We present Qwen-RobotNav, a scalablenavigation modelbuilt on Qwen-RobotNav that addresses it through aparameterised interfacewith two complementary dimensions: multipletask modesthat select the navigation behaviour, and controllableobservation parameters(e.g., token budget, per-camera weights) that govern how visual history is encoded. With training-time randomization over all parameters, Qwen-RobotNav is robust to any inference-time configuration requiring zero architectural modification to the Qwen-RobotNav backbone. We train Qwen-RobotNav on 15.6M samples; co-training withvision-language dataprevents the collapse into reactive action-sequence mappers observed intrajectory-only training. Theparameterised interfacealso makes Qwen-RobotNav a natural building block for agentic systems: for long-horizon scenarios, an upper-level planner decomposes goals into sub-tasks and dynamically switches Qwen-RobotNav’s task mode and context strategy mid-episode, composing complex behaviours from repeated calls to the same model. Extensive experiments show that Qwen-RobotNav sets new state-of-the-art results across major navigation benchmarks. The model exhibits favourable scaling from 2B to 8B parameters, with jointmulti-task trainingdeveloping a sharedspatial-planning substratethat transfers across task families, and demonstrates strong zero-shot generalisation to real-world robots across diverse environments.

View arXiv pageView PDFProject pageGitHub0Add to collection

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.18112 in a model README.md to link it from this page.

Datasets citing this paper1

#### cy0307/awesome-egocentric-atlas Viewer• Updated5 days ago • 638 • 851 • 2

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.18112 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models

Hugging Face Daily Papers

Presents Qwen-RobotManip, a Vision-Language-Action foundation model for robotic manipulation that achieves generalization through unified alignment across representation, motion, and behavior dimensions, enabling large-scale training on diverse data sources. It outperforms prior state-of-the-art models across multiple out-of-distribution benchmarks and demonstrates emergent capabilities like zero-shot instruction following and cross-embodiment transfer.

Qwen's Embodied World Modeling (28 minute read)

TLDR AI

The Qwen-RobotWorld technical report presents a unified language-conditioned video world model for embodied intelligence, enabling future video prediction from current observations across various domains like robotics, autonomous driving, and navigation, with applications in synthetic data generation, policy evaluation, and planning.

Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation

Hugging Face Daily Papers

Qwen-RobotWorld is a language-conditioned video world model that predicts future visual trajectories across multiple robotic domains using a double-stream diffusion transformer and an 8.6M video-text corpus. It unifies embodied world modeling for robotic manipulation, autonomous driving, indoor navigation, and human-to-robot transfer, achieving top benchmarks on EWMBench and DreamGen Bench.

Qwen/Qwen-AgentWorld-35B-A3B

Hugging Face Models Trending

Qwen releases Qwen-AgentWorld-35B-A3B, a native language world model that simulates agentic environments across seven domains via long chain-of-thought reasoning. The model is trained with a three-stage pipeline and supports MCP, Search, Terminal, SWE, Android, Web, and OS interactions.