TrafficAlign：面向交通场景生成的大语言模型对齐
TrafficAlign: Aligning Large Language Models for Traffic Scenario Generation
用真实驾驶视频自动合成并校验交通场景，再以这些场景对齐/微调 LLM，为自动驾驶感知模型提供可挖掘失效模式的困难样本生成管线；其「生成—验证—对齐」闭环可直接用于构造碰撞密集的测试集与训练增强数据。
原文依据 synthesizes traffic scenarios based on real-world driving videos, performs data validation, and aligns LLMs with the synthesized scenarios
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期研究已探索利用大语言模型（LLM）生成自动驾驶所需的交通场景。然而，预训练 LLM 往往难以与真实世界的交通分布对齐。在本工作中，我们提出 TrafficAlign，一个自动化框架：它基于真实驾驶视频合成交通场景，执行数据校验，并使 LLM 与合成场景对齐。评估表明，TrafficAlign 生成的交通场景高度有效，在三个自动驾驶模型上平均揭示出比最优方法多出 10.8% 的碰撞。此外，使用 TrafficAlign 生成的场景对这些驾驶模型进行微调，相比原始模型可将碰撞率显著降低 36.1%。一项使用来自六个地理上多样区域的交通数据集开展的定性研究表明，TrafficAlign 生成的场景与这些区域相应的交通分布呈现出很强的对齐。
英文原文摘要 1005 字符
Recent research has investigated the use of large language models (LLMs) to generate traffic scenarios for autonomous driving. However, pretrained LLMs often fail to align with real-world traffic distributions. In this work, we present TrafficAlign, an automated framework that synthesizes traffic scenarios based on real-world driving videos, performs data validation, and aligns LLMs with the synthesized scenarios. The evaluation shows that traffic scenarios generated by TrafficAlign are highly effective, revealing up to 10.8% more collisions on average across three autonomous driving models than state-of-the-art methods. Furthermore, fine-tuning these driving models with TrafficAlign-generated scenarios significantly reduced collision rates by 36.1% compared with the original models. A qualitative study using traffic datasets from six geographically diverse regions shows that TrafficAlign-generated scenarios exhibit strong alignment with corresponding traffic distributions in these regions.