Harnessing multimodal large language models for traffic knowledge graph generation and decision-making  

作　　者：Senyun Kuang Yang Liu Xin Wang Xinhua Wu Yintao Wei 

机构地区：[1]School of Vehicle and Mobility,Tsinghua University,Beijing,100084,China [2]Department of Architecture and Civil Engineering,Chalmers University of Technology,Gothenburg,SE-41296,Sweden [3]Department of Civil and Environmental Engineering,Northeastern University,Boston,MA,02115,USA

出　　处：《Communications in Transportation Research》2024年第1期378-381,共4页交通研究通讯(英文)

基　　金：National Natural Science Foundation of China(Grant Nos.51761135124,11672148,52003142,and 51775293).

摘　　要：1.Introduction Autonomous driving advancements have increased the importance of understanding traffic scenes for intelligent transportation systems.Substantial progress has been made in traditional tasks such as road segmentation and traffic sign recognition(Wandelt et al.,2024).However,these methods are predominantly low-level methods that focus on individual scene elements without fully addressing higher-level comprehension and decision support.

关 键 词：MODAL TRAFFIC driving 

分 类 号：TN9[电子电信—信息与通信工程]