ADS-Edit：面向自动驾驶系统的多模态知识编辑数据集

ADS-Edit: A Multimodal Knowledge Editing Dataset for Autonomous Driving Systems

March 26, 2025

作者: Chenxi Wang, Jizhan Fang, Xiang Chen, Bozhong Tian, Ziwen Xu, Huajun Chen, Ningyu Zhang

cs.AI

摘要

近期，大型多模态模型（LMMs）在自动驾驶系统（ADS）中的应用展现出显著潜力。然而，其直接应用于ADS仍面临诸多挑战，如对交通知识的误解、复杂的道路条件以及车辆状态的多样性。为应对这些挑战，我们提出采用知识编辑技术，该技术能够在不需全面重新训练的情况下，对模型行为进行针对性调整。同时，我们推出了ADS-Edit，这是一个专为ADS设计的多模态知识编辑数据集，涵盖了多种现实场景、多样化的数据类型及全面的评估指标。通过一系列详尽的实验，我们得出了若干富有启发性的结论。我们期望本工作能推动知识编辑技术在自动驾驶领域的进一步应用与发展。相关代码与数据已公开于https://github.com/zjunlp/EasyEdit。

English

Recent advancements in Large Multimodal Models (LMMs) have shown promise in Autonomous Driving Systems (ADS). However, their direct application to ADS is hindered by challenges such as misunderstanding of traffic knowledge, complex road conditions, and diverse states of vehicle. To address these challenges, we propose the use of Knowledge Editing, which enables targeted modifications to a model's behavior without the need for full retraining. Meanwhile, we introduce ADS-Edit, a multimodal knowledge editing dataset specifically designed for ADS, which includes various real-world scenarios, multiple data types, and comprehensive evaluation metrics. We conduct comprehensive experiments and derive several interesting conclusions. We hope that our work will contribute to the further advancement of knowledge editing applications in the field of autonomous driving. Code and data are available in https://github.com/zjunlp/EasyEdit.