Think before Go: Hierarchical Reasoning for Image-goal Navigation

arXiv:2604.17407 · cs.RO · Submitted 2026-04-19 · Read on arXiv

cs.RO

Submitted: 2026-04-19

Updated: 2026-10-17

Comments: Accepted by ACL2026 Oral

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Hierarchical Framework — 1× in this paper · explained in 2 episodes
• Online Reinforcement Learning — 1× in this paper · explained in 2 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• GPT-4 Technical Report
• On Evaluation of Embodied Navigation Agents
• Qwen2.5-VL Technical Report
• Matterport3D: Learning from RGB-D Data in Indoor Environments
• NaVILA: Legged Robot Vision-Language-Action Model for Navigation
• The Llama 3 Herd of Models
• Image-Goal Navigation Using Refined Feature Guidance and Scene Graph Enhancement
• MobileVLA-R1: Reinforcing Vision-Language-Action for Mobile Robots
• Adam: A Method for Stochastic Optimization
• Room-Across-Room: Multilingual Vision-and-Language Navigation with Dense Spatiotemporal Grounding
• CompassNav: Steering From Path Imitation To Decision Understanding In Navigation
• Camera-aware Label Refinement for Unsupervised Person Re-identification
• InstructNav: Zero-shot System for Generic Instruction Navigation in Unexplored Environment
• HiMemVLN: Enhancing Reliability of Open-Source Zero-Shot Vision-and-Language Navigation with Hierarchical Memory System
• PatchCue: Enhancing Vision-Language Model Reasoning with Patch-Based Visual Cues
• VLN-R1: Vision-Language Navigation via Reinforcement Fine-Tuning
• NavigateDiff: Visual Predictors are Zero-Shot Navigation Assistants
• Habitat-Matterport 3D Dataset (HM3D): 1000 Large-scale 3D Environments for Embodied AI
• Learning to Navigate Unseen Environments: Back Translation with Environmental Dropout
• StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling

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