大语言模型驱动的自动驾驶车辆在行人让行中继承人类驾驶员偏见：来自新基准的结果与启示
LLM-Driven Autonomous Vehicles Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From A New Benchmark
浏览论文内容
中文总结 AI 辅助
该研究针对LLMs和VLMs提出两种偏见测试方法，发现其行人让行决策继承人类偏见，质疑“常识”模型范式的合理性。
中文摘要 AI 辅助
公众对自动驾驶车辆（AVs）的信任不仅取决于技术成功，还取决于其决策的公平性。近期AV研究的一个趋势是使用通用“常识”模型指导AV决策，但这些模型在多大程度上继承了人类驾驶偏见仍未得到充分研究。心理学研究已表明人类驾驶员存在偏见，例如美国对黑人行人的让行率较低，因此我们认为模型偏见分析也应成为AV评估的一部分。具体而言，本文针对大语言模型（LLMs）和视觉-语言模型（VLMs）提出两种新的偏见测试方法——“其他条件相同”测试与“自我一致性”测试，以评估行人让行决策中的偏见。我们的研究结果显示，LLMs和VLMs的让行决策均受行人的性别、种族、宗教、残疾状况、年龄、肤色及社会经济地位影响。尽管不同模型的偏见类型和程度存在差异，但我们突出了常见模式，并对“常识”模型范式提出质疑，特别是需要修订该范式或解决下游偏见问题。
英文摘要
Public trust in Autonomous Vehicles (AVs) may depend not only on technical success but also on the fairness of their decision making. While a recent trend in AV research involves using general purpose "common sense" models to guide AV decision making, the degree to which these inherit human biases in driving is still understudied. Given that psychology studies have shown human driver biases exist, such as lower pedestrian-yielding rates to Black pedestrians in the US, we argue that analyses of model bias should also be part of AV evaluation. Concretely, in this paper we propose two new bias testing methodologies for Large Language Models (LLMs) and Visual-Language Models (VLMs)-"All Else Being Equal" tests and "Self-Consistency" tests-in order to assess bias in pedestrian-yielding decisions. Our findings show that both LLMs and VLMs make yielding decisions which are influenced by pedestrian gender, ethnicity, religion, disability, age, skin tone and socio-economic status. While the type and degree of bias is different from model to model, we highlight common patterns-and raise questions about the "common sense" model paradigm, particularly the need to either revise the paradigm or address issues of downstream bias.
