paper-with-me

홈 › Papers

WAM-Flow: Parallel Coarse-to-Fine Motion Planning via Discrete Flow Matching for Autonomous Driving

2025-12-05 · Yifang Xu, Jiahao Cui, Feipeng Cai, Zhihao Zhu, Hanlin Shang, Shan Luan, Mingwang Xu, Neng Zhang, Yaoyi Li, Jia Cai, Siyu Zhu arxiv

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 89.1 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. The code will be publicly available soon.

📄 PDF Abstract BibTeX arXiv:2512.06112

Code (0)

등록된 구현이 없습니다.

Tasks

Trajectory PlanningAutonomous DrivingMotion Planning

Similar Papers 제목 키워드 기반

Parallelised Diffeomorphic Sampling-based Motion Planning

2021-08-26 · Tin Lai, Weiming Zhi, Tucker Hermans, Fabio Ramos

We propose Parallelised Diffeomorphic Sampling-based Motion Planning (PDMP). PDMP is a novel parallelised framework that uses bijective and differentiable mappings, or diffeomorphisms, to transform sampling distributions…

MORPHMotion PlanningNormalising Flowsvalid

PlanMoGPT: Flow-Enhanced Progressive Planning for Text to Motion Synthesis

2025-06-22 · Chuhao Jin, Haosen Li, Bingzi Zhang, Che Liu 외

Recent advances in large language models (LLMs) have enabled breakthroughs in many multimodal generation tasks, but a significant performance gap still exists in text-to-motion generation, where LLM-based methods lag far…

DiversityMotion GenerationMotion Synthesismultimodal generation

EmoInstruct-TTS: Dual-Path Instruction-Guided Emotional Speech Synthesis

2026-06-08 · Minghui Wu, Ganjun Liu, Zikun Fang, Ting Meng 외 arxiv

Instruction-based controllable speech synthesis enables users to specify emotions through natural language. However, existing approaches often rely on coarse emotion labels and lack explicit modeling of fine-grained inte…

Speech Synthesis

Parareal with a Learned Coarse Model for Robotic Manipulation

2019-12-12 · Wisdom Agboh, Oliver Grainger, Daniel Ruprecht, Mehmet Dogar

A key component of many robotics model-based planning and control algorithms is physics predictions, that is, forecasting a sequence of states given an initial state and a sequence of controls. This process is slow and a…

MuJoCo

MPC-based Coarse-to-Fine Motion Planning for Robotic Object Transportation in Cluttered Environments

2025-07-15 · Chen Cai, Ernesto Dickel Saraiva, Ya-jun Pan, Steven Liu arxiv

This letter presents a novel coarse-to-fine motion planning framework for robotic manipulation in cluttered, unmodeled environments. The system integrates a dual-camera perception setup with a B-spline-based model predic…

Motion Planning