SeeNav-Agent

SeeNav-Agent: Enhancing Vision-Language Navigation with Visual Prompt and Step-Level Policy Optimization

This repository contains the official implementation for the paper SeeNav-Agent: Enhancing Vision-Language Navigation with Visual Prompt and Step-Level Policy Optimization.

<div align="center"> <a href="https://github.com/WzcTHU/SeeNav-Agent"><img src="https://img.shields.io/badge/GitHub-Code-blue.svg?logo=github&" alt="GitHub Code"></a> <a href="https://huggingface.co/wangzc9865/SeeNav-Agent"><img src="https://img.shields.io/badge/🤗&nbsp;-HuggingFace-blue" alt="Hugging Face Model"></a> </div>

Overview

We propose SeeNav-Agent, a novel LVLM-based embodied navigation framework that includes a zero-shot dual-view visual prompt technique for the input side and an efficient RFT algorithm named SRGPO for post-training. Existing Vision-Language Navigation (VLN) agents often suffer from perception, reasoning, and planning errors, which SeeNav-Agent aims to mitigate through its proposed techniques.

🚀 Highlights

• 🚫 Zero-Shot Visual Prompt: No extra training for performance improvement with visual prompt.
• 🗲 Efficient Step-Level Advantage Calculation: Step-Level groups are randomly sampled from the entire batch.
• 📈 Significant Gains: +20.0pp (GPT4.1+VP) and +5.6pp (Qwen2.5-VL-3B+VP+SRGPO) improvements on EmbodiedBench-Navigation.

📖 Summary

<div style="text-align: center;"> <img src="https://github.com/WzcTHU/SeeNav-Agent/raw/main/SeeNav/figs/framework.png" width="100%"> </div>

• 🎨 Dual-View Visual Prompt: We apply visual prompt techniques directly on the input dual-view image to reduce the visual hallucination.
• 🔁 Step Reward Group Policy Optimization (SRGPO): By defining a state-independent verifiable process reward function, we achieve efficient step-level random grouping and advantage estimation.

📋 Results on EmbodiedBench-Navigation

📝 Main Results

<div align="center"> <img src="https://github.com/WzcTHU/SeeNav-Agent/raw/main/SeeNav/figs/results.png" width="50%"/> </div>

🖌️ Training Curves for RFT

<div align="center"> <img src="https://github.com/WzcTHU/SeeNav-Agent/raw/main/SeeNav/figs/training_curves.png" width="50%"/> </div>

🖍️ Testing Curves for OOD-Scenes

<div align="center"> <img src="https://github.com/WzcTHU/SeeNav-Agent/raw/main/SeeNav/figs/ood_val.png" width="50%"/> </div>

📦 Checkpoint

🛠️ Usage

Setup

1. Setup a seperate environment for evaluation according to: EmbodiedBench-Nav and Qwen3-VL to support Qwen2.5-VL-3B-Instruct.

2. Setup a seperate training environment according to: verl-agent and Qwen3-VL to support Qwen2.5-VL-3B-Instruct.

Evaluation

Use the following command to evaluate the model on EmbodiedBench:

conda activate <your_env_for_eval>
cd SeeNav
python testEBNav.py

Hint: you need to first set your endpoint, API-key and api_version in SeeNav/planner/models/remote_model.py

Training

verl-agent/examples/srgpo_trainer contains example scripts for SRGPO-based training on EmbodiedBench-Navigation.

1. Modify run_ebnav.sh according to your setup.

2. Run the following command:

conda activate <your_env_for_train>
cd verl-agent
bash examples/srgpo_trainer/run_ebnav.sh

📚 Citation

If you find this work helpful in your research, please consider citing:

@article{wang2025seenav,
  title={SeeNav-Agent: Enhancing Vision-Language Navigation with Visual Prompt and Step-Level Policy Optimization},
  author={Zhengcheng Wang and Zichuan Lin and Yijun Yang and Haobo Fu and Deheng Ye},
  journal={arXiv preprint arXiv:2512.02631},
  year={2025}
}