arXiv 2602.145772026-02-16cs.CV

DriveFine: 通过细化增强的掩码扩散VLA实现精确且鲁棒的驾驶

DriveFine: Refining-Augmented Masked Diffusion VLA for Precise and Robust Driving

Huazhong University of Science and Technology（华中科技大学）；Xiaomi EV（小米电动车）；Institute for AI Industry Research (AIR), Tsinghua University（人工智能产业研究院（AIR），清华大学）

Chenxu Dang, Sining Ang, Yongkang Li, Haochen Tian, Jie Wang, Guang Li, Hangjun Ye, Jie Ma, Long Chen, Yan Wang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

DriveFine提出一种结合灵活解码与自我纠正能力的掩码扩散VLA模型，通过插件式块MoE和混合强化学习策略提升自动驾驶规划的精度与鲁棒性。

摘要：

视觉-语言-动作（VLA）模型在自动驾驶中越来越多地采用通过模仿学习训练后接强化学习的生成规划器。基于扩散的规划器面临模态对齐困难、训练效率低和泛化能力有限的问题。基于标记的规划器则受累积因果误差和不可逆解码的困扰。总而言之，两种主流范式具有互补的优势和劣势。在本文中，我们提出了DriveFine，一种结合灵活解码和自我纠正能力的掩码扩散VLA模型。特别是，我们设计了一个新颖的插件式块MoE，它无缝地在生成专家之上注入一个细化专家。通过在推理过程中启用显式专家选择并在训练过程中阻断梯度，两个专家完全解耦，保留了预训练权重的基础能力和通用模式，突显了块MoE设计的灵活性和可扩展性。此外，我们设计了一种混合强化学习策略，该策略鼓励有效探索细化专家的同时保持训练稳定性。在NAVSIM v1、v2和Navhard基准上的大量实验表明，DriveFine表现出强大的效果和鲁棒性。代码将在https://github.com/MSunDYY/DriveFine上发布。

英文摘要：

Vision-Language-Action (VLA) models for autonomous driving increasingly adopt generative planners trained with imitation learning followed by reinforcement learning. Diffusion-based planners suffer from modality alignment difficulties, low training efficiency, and limited generalization. Token-based planners are plagued by cumulative causal errors and irreversible decoding. In summary, the two dominant paradigms exhibit complementary strengths and weaknesses. In this paper, we propose DriveFine, a masked diffusion VLA model that combines flexible decoding with self-correction capabilities. In particular, we design a novel plug-and-play block-MoE, which seamlessly injects a refinement expert on top of the generation expert. By enabling explicit expert selection during inference and gradient blocking during training, the two experts are fully decoupled, preserving the foundational capabilities and generic patterns of the pretrained weights, which highlights the flexibility and extensibility of the block-MoE design. Furthermore, we design a hybrid reinforcement learning strategy that encourages effective exploration of refinement expert while maintaining training stability. Extensive experiments on NAVSIM v1, v2, and Navhard benchmarks demonstrate that DriveFine exhibits strong efficacy and robustness. The code will be released at https://github.com/MSunDYY/DriveFine.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。