Despite rapid progress in autonomous driving, reliable training and evaluation of driving systems remain fundamentally constrained by the lack of scalable and interactive simulation environments. Recent generative video models achieve remarkable visual fidelity, yet most operate in open-loop settings and fail to support fine-grained frame-level interaction between agent actions and environment evolution. Building a learning-based closed-loop simulator for autonomous driving poses three major challenges: maintaining long-horizon temporal and cross-view consistency, mitigating autoregressive degradation under iterative self-conditioning, and satisfying low-latency inference constraints. In this work, we propose FAR-Drive, a frame-level autoregressive video generation framework for autonomous driving. We introduce a multi-view diffusion transformer with fine-grained structured control, enabling geometrically consistent multi-camera generation. To address long-horizon consistency and iterative degradation, we design a two-stage training strategy consisting of adaptive reference horizon conditioning and blend-forcing autoregressive training, which progressively improves consistency and robustness under self-conditioning. To meet low-latency interaction requirements, we further integrate system-level efficiency optimizations for inference acceleration. Experiments on the nuScenes dataset demonstrate that our method achieves state-of-the-art performance among existing closed-loop autonomous driving simulation approaches, while maintaining sub-second latency on a single GPU.

翻译：尽管自动驾驶领域进展迅速，但驾驶系统的可靠训练与评估仍因缺乏可扩展且交互式的仿真环境而受到根本性制约。近期的生成式视频模型取得了显著的视觉保真度，但大多在开环设置下运行，无法支持智能体动作与环境演化之间的细粒度帧级交互。构建基于学习的闭环自动驾驶仿真器面临三大挑战：保持长时域时间与跨视角一致性，缓解迭代自条件化下的自回归退化，以及满足低延迟推理约束。本文提出FAR-Drive，一种面向自动驾驶的帧级自回归视频生成框架。我们引入具有细粒度结构化控制的多视角扩散Transformer，实现几何一致的多摄像头生成。针对长时域一致性与迭代退化问题，我们设计了包含自适应参考时域条件化与混合强制自回归训练的两阶段训练策略，逐步提升自条件化下的模型一致性与鲁棒性。为满足低延迟交互需求，我们进一步集成系统级效率优化以实现推理加速。在nuScenes数据集上的实验表明，本方法在现有闭环自动驾驶仿真方案中达到最优性能，同时在单GPU上保持亚秒级延迟。