ETP-R1: Evolving Topological Planning with Reinforcement Fine-tuning for Vision-Language Navigation in Continuous Environments
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918262764208128 |
|---|---|
| author | Ye, Shuhao Mao, Sitong Cui, Yuxiang Yu, Xuan Zhai, Shichao Chen, Wen Zhou, Shunbo Xiong, Rong Wang, Yue |
| author_facet | Ye, Shuhao Mao, Sitong Cui, Yuxiang Yu, Xuan Zhai, Shichao Chen, Wen Zhou, Shunbo Xiong, Rong Wang, Yue |
| contents | Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to navigate towards target in continuous environments, following natural language instructions. While current graph-based methods offer an efficient, structured approach by abstracting the environment into a topological map and simplifying the action space to waypoint selection, they lag behind methods based on Large Vision-Language Models (LVLMs) in leveraging large-scale data and advanced training paradigms. In this paper, we try to bridge this gap by introducing ETP-R1, a framework that applies the paradigm of scaling up data and Reinforcement Fine-Tuning (RFT) to a graph-based VLN-CE model. To build a strong foundation, we first construct a high-quality, large-scale pretraining dataset using the Gemini API. This dataset consists of diverse, low-hallucination instructions for topological trajectories, providing rich supervision for our graph-based policy to map language to topological paths. This foundation is further strengthened by unifying data from both R2R and RxR tasks for joint pretraining. Building on this, we introduce a three-stage training paradigm, which culminates in the first application of closed-loop, online RFT to a graph-based VLN-CE model, powered by the Group Relative Policy Optimization (GRPO) algorithm. Extensive experiments demonstrate that our approach is highly effective, establishing new state-of-the-art performance across all major metrics on both the R2R-CE and RxR-CE benchmarks. Our code is available at https://github.com/Cepillar/ETP-R1. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_20940 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ETP-R1: Evolving Topological Planning with Reinforcement Fine-tuning for Vision-Language Navigation in Continuous Environments Ye, Shuhao Mao, Sitong Cui, Yuxiang Yu, Xuan Zhai, Shichao Chen, Wen Zhou, Shunbo Xiong, Rong Wang, Yue Robotics Vision-Language Navigation in Continuous Environments (VLN-CE) requires an embodied agent to navigate towards target in continuous environments, following natural language instructions. While current graph-based methods offer an efficient, structured approach by abstracting the environment into a topological map and simplifying the action space to waypoint selection, they lag behind methods based on Large Vision-Language Models (LVLMs) in leveraging large-scale data and advanced training paradigms. In this paper, we try to bridge this gap by introducing ETP-R1, a framework that applies the paradigm of scaling up data and Reinforcement Fine-Tuning (RFT) to a graph-based VLN-CE model. To build a strong foundation, we first construct a high-quality, large-scale pretraining dataset using the Gemini API. This dataset consists of diverse, low-hallucination instructions for topological trajectories, providing rich supervision for our graph-based policy to map language to topological paths. This foundation is further strengthened by unifying data from both R2R and RxR tasks for joint pretraining. Building on this, we introduce a three-stage training paradigm, which culminates in the first application of closed-loop, online RFT to a graph-based VLN-CE model, powered by the Group Relative Policy Optimization (GRPO) algorithm. Extensive experiments demonstrate that our approach is highly effective, establishing new state-of-the-art performance across all major metrics on both the R2R-CE and RxR-CE benchmarks. Our code is available at https://github.com/Cepillar/ETP-R1. |
| title | ETP-R1: Evolving Topological Planning with Reinforcement Fine-tuning for Vision-Language Navigation in Continuous Environments |
| topic | Robotics |
| url | https://arxiv.org/abs/2512.20940 |