HelloWorld：迈向生成式驾驶世界模型的实际应用
HelloWorld: Towards Practical Applications of Generative Driving World Models
AI总结：
本文提出HelloWorld，一个20亿参数的驾驶世界模型系统，通过特化视觉运动先验并采用块因果生成接口，实现可控、多传感器、高效的驾驶数据生成与交互式模拟。
AI中文摘要：
驾驶世界模型为可扩展的反事实数据生成和超越记录驾驶日志的交互式模拟提供了一条有前景的途径。实现这一潜力需要一个能够泛化到多样场景、忠实响应规定控制、生成连贯的多传感器观测，并在重复推理下高效运行的系统。我们提出了HelloWorld，一个围绕这些需求设计的20亿参数驾驶世界模型系统。HelloWorld通过利用自车位姿、高清地图和3D边界框，逐步将来自异构视频数据的广泛视觉和运动先验特化为可控的驾驶生成。一个块因果生成接口，结合对自生成上下文的适应，使模型与顺序模拟对齐。该系统进一步支持同步的七摄像头RGB生成和有条件的LiDAR合成，并被蒸馏到少步推理以实现高效部署。实验评估了视觉质量、控制保真度、跨视角一致性、重复生成下的鲁棒性、推理效率和LiDAR合成。总之，HelloWorld为可扩展的驾驶数据生成和交互式模拟提供了一个统一框架。
英文摘要：
Driving world models provide a promising route toward scalable counterfactual data generation and interactive simulation beyond recorded driving logs. Realizing this potential requires a system that can generalize across diverse scenes, respond faithfully to prescribed controls, generate coherent multi-sensor observations, and operate efficiently under repeated inference. We present \textbf{HelloWorld}, a 2B driving world model system designed around these requirements. HelloWorld progressively specializes broad visual and motion priors from heterogeneous video data into controllable driving generation using ego pose, HD maps, and 3D boxes. A block-causal generation interface, together with adaptation to self-generated context, aligns the model with sequential simulation. The system further supports synchronized seven-camera RGB generation and conditional LiDAR synthesis, and is distilled toward few-step inference for efficient deployment. Experiments evaluate visual quality, control fidelity, cross-view consistency, robustness under repeated generation, inference efficiency, and LiDAR synthesis. Together, HelloWorld provides a unified framework for scalable driving data generation and interactive simulation.
