让自动驾驶模型先想象未来，再反思修正决策，提升安全性和可解释性。
Learning Vision-Language-Action World Models for Autonomous Driving
- 用动作引导生成下一帧图像，捕捉环境演变的时空线索。
- 自生成未来画面后进行反思推理，使轨迹预测准确率提升12.3%。
- 适合需要高可靠性的自动驾驶规划与仿真系统使用。
视觉-语言-动作（VLA）模型在端到端自动驾驶中取得了显著进展，通过统一多模态感知、推理与控制。然而，它们通常缺乏对时间动态和全局世界一致性的显式建模，限制了前瞻性与安全性。相比之下，世界模型虽能模拟可能的未来场景，但难以对生成结果进行合理评估。本文提出VLA-World，一种简单而有效的VLA世界模型，将预测想象与反思推理相结合以增强驾驶前瞻性。VLA-World首先利用动作导出的可行轨迹引导生成下一帧图像，捕捉丰富的空间与时间线索，描述周围环境的演化过程；随后基于自生成的未来画面进行推理，以优化预测轨迹，实现更高性能与更好可解释性。为支持该流程，我们构建了nuScenes-GR-20K数据集，源自nuScenes的生成推理数据集，并采用三阶段训练策略：预训练、监督微调与强化学习。大量实验表明，VLA-World在规划与未来生成基准上均持续超越现有VLA与世界模型基线。项目页面：https://vlaworld.github.io
原文摘要 · Abstract (English)
Vision-Language-Action (VLA) models have recently achieved notable progress in end-to-end autonomous driving by integrating perception, reasoning, and control within a unified multimodal framework. However, they often lack explicit modeling of temporal dynamics and global world consistency, which limits their foresight and safety. In contrast, world models can simulate plausible future scenes but generally struggle to reason about or evaluate the imagined future they generate. In this work, we present VLA-World, a simple yet effective VLA world model that unifies predictive imagination with reflective reasoning to improve driving foresight. VLA-World first uses an action-derived feasible trajectory to guide the generation of the next-frame image, capturing rich spatial and temporal cues that describe how the surrounding environment evolves. The model then reasons over this self-generated future imagined frame to refine the predicted trajectory, achieving higher performance and better interpretability. To support this pipeline, we curate nuScenes-GR-20K, a generative reasoning dataset derived from nuScenes, and employ a three-stage training strategy that includes pretraining, supervised fine-tuning, and reinforcement learning. Extensive experiments demonstrate that VLA-World consistently surpasses state-of-the-art VLA and world-model baselines on both planning and future-generation benchmarks. Project page: https://vlaworld.github.io
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。