DriveCombo：自动驾驶中组合式交通规则推理的基准评测
DriveCombo: Benchmarking Compositional Traffic Rule Reasoning in Autonomous Driving
为视觉语言模型在自动驾驶中的规则理解提供了分层认知评测协议与 Rule2Scene 规则到场景的生成管线，可直接用于诊断 MLLM 在多规则冲突下的推理失效，并作为微调数据来源提升下游规划。
原文依据 we propose DriveCombo, a text and vision-based benchmark for compositional traffic rule reasoning
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论多模态大语言模型（MLLM）正迅速成为端到端自动驾驶系统的智能大脑。一个关键挑战在于评估 MLLM 是否能够真正理解并遵循复杂的真实世界交通规则。然而，现有基准主要聚焦于交通标志识别等单一规则场景，忽视了真实驾驶中多规则并发与冲突的复杂性。因此，模型在简单任务上表现良好，却在真实世界的复杂情境中经常失败或违反规则。为弥补这一空白，我们提出 DriveCombo，一个面向组合式交通规则推理的文本与视觉基准。受人类驾驶员认知发展的启发，我们提出系统化的五级认知阶梯，从单一规则理解到多规则整合与冲突消解来评估推理能力，从而实现对不同认知阶段的定量评估。我们进一步提出 Rule2Scene Agent，通过规则构造与场景生成将基于语言的交通规则映射到动态驾驶场景，从而实现场景级的交通规则视觉推理。对 14 个主流 MLLM 的评估显示，随着任务难度提升，性能出现明显下降，尤其是在规则冲突消解任务中。在划分数据集并在训练集上微调之后，我们进一步观察到交通规则推理与下游规划能力均有显著提升。这些结果凸显了 DriveCombo 在推进合规且智能的自动驾驶系统方面的有效性。
英文原文摘要 1575 字符
Multimodal Large Language Models (MLLMs) are rapidly becoming the intelligence brain of end-to-end autonomous driving systems. A key challenge is to assess whether MLLMs can truly understand and follow complex real-world traffic rules. However, existing benchmarks mainly focus on single-rule scenarios like traffic sign recognition, neglecting the complexity of multi-rule concurrency and conflicts in real driving. Consequently, models perform well on simple tasks but often fail or violate rules in real world complex situations. To bridge this gap, we propose DriveCombo, a text and vision-based benchmark for compositional traffic rule reasoning. Inspired by human drivers' cognitive development, we propose a systematic Five-Level Cognitive Ladder that evaluates reasoning from single-rule understanding to multi-rule integration and conflict resolution, enabling quantitative assessment across cognitive stages. We further propose a Rule2Scene Agent that maps language-based traffic rules to dynamic driving scenes through rule crafting and scene generation, enabling scene-level traffic rule visual reasoning. Evaluations of 14 mainstream MLLMs reveal a pronounced decline in performance as task difficulty increases, especially in the rule conflict resolution task. After splitting the dataset and fine-tuning on the training set, we further observe substantial improvements in both traffic rule reasoning and downstream planning capabilities. These results highlight the effectiveness of DriveCombo in advancing compliant and intelligent autonomous driving systems.