跳到正文
arXiv cs.AI· Baiting Luo, Shatadal Mishra, Ahmadreza Moradipari, Nejib Ammar·· 4 小时前AI 评分34

AR-WAM:面向自动驾驶的高效流式世界-动作模型

AR-WAM: efficient streaming world-action modeling for autonomous driving

AI 导读

AR-WAM 是一种面向自动驾驶的高效流式世界-动作模型,将自回归时序建模与单次前向的未来条件化结合,其因果视频骨干可增量更新视觉上下文并复用历史计算。仅用单个前摄像头,AR-WAM 在 NAVSIM-v1 上监督训练后达到 91.5 PDMS,经强化学习后提升至 94.0,优于对比的强端到端世界-动作模型。

正文 · 原文

View PDF HTML (experimental)

Abstract:Autonomous driving requires anticipating how traffic will evolve while responding quickly to new observations. Video foundation models offer the first ability, but repeated history encoding and iterative video generation make them costly to use for driving. We present AR-WAM, an efficient streaming world-action model that combines autoregressive temporal modeling with single-pass future conditioning. Because its video backbone is causal, AR-WAM updates its visual context incrementally, reusing past computation and conditioning future prediction on observed motion. We further hypothesize that future features computed from recent observations can improve generalization in dynamic traffic without iterative video generation. AR-WAM therefore obtains these features in one forward pass and uses them to condition trajectory generation, and reinforcement learning then improves the policy through direct optimization of driving rewards. With a single front camera, AR-WAM achieves 91.5 PDMS after supervised training and 94.0 after reinforcement learning on NAVSIM-v1, outperforming the compared strong end-to-end world-action models. In a dynamic setting of continuous driving, supervised AR-WAM keeps its visual history across decisions and outperforms strong world-action baselines.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
Cite as: arXiv:2611.21851 [cs.CV]
  (or arXiv:2611.21851v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2611.21851

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Baiting Luo [view email]
[v1] Mon, 16 Nov 2026 20:16:08 UTC (7,390 KB)

来源:arXiv cs.AI · arxiv.org