X-NavDP: Generalizing Navigation Diffusion Policy to Novel Behavior and Embodiments with Group Q-score Reweighted Matching

X-NavDP: Generalizing Navigation Diffusion Policy to Novel Behavior and Embodiments with Group Q-score Reweighted Matching

|Authors: Tianyu Yang, Yiming Zeng, Wenzhe Cai, Yuqiang Yang, Jiaqi Peng, Hui Cheng, Jiangmiao Pang, Tai Wang

|Venue: arXiv preprint | Year: 2026

|arXiv: 2607.28560v1

Abstract

Pretraining navigation diffusion policies rely on large-scale expert demonstrations. These data are typically generated by a fully-informed oracle planner suited to a single nominal robot. This limits the policy’s generalization to diverse embodiments and challenging scenarios (e.g., escaping dead ends or detouring long obstacles) that demand diverse local reactive behaviors with only onboard local observations. Post-training the policy with reinforcement learning (RL) offers a principled remedy. However, previous RL for diffusion approaches lead to only marginal improvements. This is because the intractable likelihood of diffusion policies renders policy gradients unstable in addition to inefficient policy exploration. To address these challenges, we propose a data-efficient diffusion RL post-training framework - GQRM (Group Q-score Reweighted Matching). Our framework introduces two complementary designs: (i) a self-bootstrapped exploration strategy with behavior perturbation that preserves the pretrained policy prior, and (ii) a group Q-score normalization mechanism that computes per-trajectory values on each state for efficient reweighted score matching. By conducting distributed online RL training across heterogeneous embodiments, the resulting fine-tuned policy, X-NavDP, achieves state-of-the-art cross-embodiment visual navigation performance, improving the overall success rate from 61.20% to 84.28% in simulation and 10% to 65% in real-world hard cases. The code and model are publicly available at https://yty-sky.github.io/x-navdp-project-page.

Key Contributions

• Pretraining navigation diffusion policies rely on large-scale expert demonstrations
• These data are typically generated by a fully-informed oracle planner suited to a single nominal …
• This limits the policy’s generalization to diverse embodiments and challenging scenarios (e.g., e…
• Post-training the policy with reinforcement learning (RL) offers a principled remedy
• However, previous RL for diffusion approaches lead to only marginal improvements

Topics

• vision
• reinforcement-learning
• planning
• learning-from-demonstration

Code & Data

• GitHub repository: https://github.com/InternRobotics/NavDP/tree/master/baselines/x-navdp

BibTeX

@article{Yang2026_260728560v1,
  title     = {X-NavDP: Generalizing Navigation Diffusion Policy to Novel Behavior and Embodiments with Group Q-score Reweighted Matching},
  author    = {Tianyu Yang and Yiming Zeng and Wenzhe Cai and Yuqiang Yang and Jiaqi Peng and Hui Cheng and Jiangmiao Pang and Tai Wang},
  year      = {2026},
  eprint    = {2607.28560v1},
  archivePrefix = {arXiv},
  primaryClass  = {cs.RO},
  url       = {https://arxiv.org/abs/2607.28560v1}
}