WAM-Flow：基于离散流匹配的并行粗到细运动规划用于自动驾驶
WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving
将离散流匹配引入端到端自动驾驶轨迹规划，用并行双向去噪替代自回归解码，实现可调计算量的粗到细规划；其度量对齐分词器和几何感知流目标可迁移到其他需要结构化输出的视觉任务。
原文依据 WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 WAM-Flow，一种视觉-语言-动作（VLA）模型，将自车轨迹规划转化为结构化词元空间上的离散流匹配。与自回归解码器不同，WAM-Flow 执行完全并行、双向去噪，支持粗到细的细化，并具有可调的计算-精度权衡。具体而言，该方法结合了通过三元组边际学习保持标量几何的度量对齐数值分词器、几何感知流目标以及模拟器引导的 GRPO 对齐，后者整合安全性、自车进度和舒适性奖励，同时保持并行生成。多阶段适配将预训练的自回归骨干网络（Janus-1.5B）从因果解码转换为非因果流模型，并通过持续多模态预训练增强道路场景能力。得益于一致性模型训练和并行解码推理的固有特性，WAM-Flow 在闭环性能上优于自回归和基于扩散的 VLA 基线，在 NAVSIM v1 基准上，1 步推理达到 88.7 PDMS，5 步推理达到 90.3 PDMS。这些结果确立了离散流匹配作为端到端自动驾驶新范式的潜力。
英文原文摘要 1256 字符
We introduce WAM-Flow, a vision-language-action (VLA) model that casts ego-trajectory planning as discrete flow matching over a structured token space. In contrast to autoregressive decoders, WAM-Flow performs fully parallel, bidirectional denoising, enabling coarse-to-fine refinement with a tunable compute-accuracy trade-off. Specifically, the approach combines a metric-aligned numerical tokenizer that preserves scalar geometry via triplet-margin learning, a geometry-aware flow objective and a simulator-guided GRPO alignment that integrates safety, ego progress, and comfort rewards while retaining parallel generation. A multi-stage adaptation converts a pre-trained auto-regressive backbone (Janus-1.5B) from causal decoding to non-causal flow model and strengthens road-scene competence through continued multimodal pretraining. Thanks to the inherent nature of consistency model training and parallel decoding inference, WAM-Flow achieves superior closed-loop performance against autoregressive and diffusion-based VLA baselines, with 1-step inference attaining 88.7 PDMS and 5-step inference reaching 90.3 PDMS on NAVSIM v1 benchmark. These results establish discrete flow matching as a new promising paradigm for end-to-end autonomous driving.