《交通运输工程与信息学报》编辑部;
【目标】启发和推动大语言模型在道路交通领域中的创新与应用。【背景】以ChatGPT为代表的大语言模型的出现深刻改变了人类社会，推动了人工智能技术的发展，并显著影响了社会互动的方式，也包括道路交通领域。【方法】首先，简要介绍了大语言模型及其主要特征；其次，列举其在道路交通领域的典型应用，梳理共性特征；最后，论述了大语言模型的一些局限性。【结果】大语言模型可以在如下几方面赋能道路交通研究与应用：减少用户与结果之间的技术障碍、帮助模型适应实际需求、交通视频自动理解、减轻梳理文本工作的负担、助力自动驾驶，换个角度讲，凡是遇到上述技术障碍，均可考虑使用大语言模型来辅助问题的解决。同时，指出了目前大语言模型的一些局限性，包括：可重复性问题、缺乏领域知识、处理效率问题、模态表征差异、“幻觉”问题、缺乏对物理世界的理解、隐私和安全问题等。
| 958 | 0 | 65 |
| 下载次数 | 被引频次 | 阅读次数 |
[1] WANG S Y, ZHU Y X, LI Z H, et al. ChatGPT as your vehicle co-pilot:an initial attempt[J]. IEEE Transactions on Intelligent Vehicles, 2023, 8(12):4706-4721.
[2] ZHANG S Y, FU D C, LIANG W Z, et al. TrafficGPT:Viewing, processing and interacting with traffic foundation models[J]. Transport Policy, 2024, 150:95-105.
[3] ZHANG K P, ZHOU F, WU L, et al. Semantic understanding and prompt engineering for large-scale traffic data imputation[J]. Information Fusion, 2024, 102:102038.
[4] DA L C, GAO M Q, MEI H, et al. Prompt to transfer:simto-real transfer for traffic signal control with prompt learning[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2024, 38(1):82-90.
[5] QASEMI E, FRANCIS J M, OLTRAMARI A. Traffic-domain video question answering with automatic captioning[EB/OL].(2023-07-18)[2024-11-06]. http://arxiv.org/abs/2307.09636v1
[6] SAKAINO H, PHUONG T N, DUY V N. PV-Cap:3D dynamic scene understanding through open physics-based vocabulary[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops(CVPRW).Seattle a:IEEE, 2024:7932-7942.
[7] XU L, HUANG H, LU J. SUTD-TrafficQA:a question answering benchmark and an efficient network for video reasoning over traffic events[EB/OL].(2021-05-29)[2024-11-05]https://doi. org/10.48550/arXiv. 2103.15538. 2021:9878-9888.
[8] WANG LN, REN Y L, JIANG H, et al. AccidentGPT:Accident analysis and prevention from V2X environmental perception with multi-modal large model[EB/OL].(2023-12-20)[2024-11-05]. https://doi. org/10.48550/arXiv.2312.13156
[9] LEONG M, ABDELHALIM A, HA J, et al. MetRoBERTa:leveraging traditional customer relationship management data to develop a transit-topic-aware language model[EB/OL].(2023-09-09)[2024-11-06]. https://doi. org/10.48550/arXiv.2308.05012
[10] ZHU Y X, WANG S Y, ZHONG W Q, et al. A Survey on large language models-empowered for autonomous driving[EB/OL].(2024-09-21)[2024-11-06]. https://arxiv.org/abs/2409.14165 2024.
[11] CAO X, ZHOU T, MA Y, et al. MAPLM:a real-world large-scale vision-language benchmark for map and traffic scene understanding[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle:IEEE, 2024:21819-21830.
基本信息:
DOI：10.19961/j.cnki.1672-4747.2024.12.001
中图分类号:TP18;U495
引用信息:
[1]贺正冰.大语言模型在道路交通领域应用：创新与挑战[J].交通运输工程与信息学报,2025,23(01):85-92.DOI:10.19961/j.cnki.1672-4747.2024.12.001.
基金信息: