Large language model empowered smart city mobility

Abstract

Smart city mobility faces mounting challenges as urban mobility systems grow increasingly complex. Large language models (LLMs) have promise in interpreting and processing multi-modal urban data, but issues like model instability, computational inefficiency, and concerns about reliability hinder their implementations. In this Comment, we outline feasible LLM application scenarios, critically evaluate existing challenges, and highlight avenues for advancing LLM-based mobility systems through multi-modal data integration and developing robust, lightweight models.

Graphical abstract

Keywords

smart city mobility / large language model / urban computing / transportation

References

[1] ChenJLuS (2024). An advanced driving agent with the multimodal large language model for autonomous vehicles. In: Proceedings of 2024 IEEE International Conference on Mobility, Operations, Services and Technologies (MOST), Dallas, USA, 1–11

[2] Chen Y, Chen X M, Gao Z, (2024). Toward equitable, transparent, and collaborative human mobility computing for smart cities. Innovation, 5( 5): 100672

[3] CuiCMaYCaoXYeWZhouYLiangKChenJLuJYangZLiaoKGaoTLiETangKCaoZZhouTLiuAYanXMeiSCaoJWangZZhengC (2024). A survey on multimodal large language models for autonomous driving. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, USA, 958–979

[4] Dong X, Yu Z, Cao W, Shi Y, Ma Q, (2020). A survey on ensemble learning. Frontiers of Computer Science, 14: 241–258

[5] DsouzaATempelmeierNYuRGottschalkSDemidovaE (2021). Worldkg: A world-scale geographic knowledge graph. In: Proceedings of the 30th ACM International Conference on Information & Knowledge Management, New York, USA, 4475–4484

[6] FangBYangZWangSDiX (2024). TraveLLM: Could you plan my new public transit route in face of a network disruption? doi: 10.48550/arXiv.2407.14926

[7] Fawzy D, Moussa S M, Badr N L, (2023). An IoT-based resource utilization framework using data fusion for smart environments. Internet of Things, 21: 100645

[8] Gao C, Lan X, Li N, Yuan Y, Ding J, Zhou Z, Xu F, Li Y, (2024). Large language models empowered agent-based modeling and simulation: a survey and perspectives. Humanities & Social Sciences Communications, 11: 1259

[9] Karger E, Jagals M, Ahlemann F, (2021). Blockchain for smart mobility—literature review and future research agenda. Sustainability, 13( 23): 13268

[10] Li Q, Wen Z, Wu Z, Hu S, Wang N, Li Y, Liu X, He B, (2021). A survey on federated learning systems: Vision, hype and reality for data privacy and protection. IEEE Transactions on Knowledge and Data Engineering, 35( 4): 3347–3366

[11] LiWGaoCNiuGXiaoXLiuHLiuJWuHWangH (2020). Unimo: Towards unified-modal understanding and generation via cross-modal contrastive learning. doi: 10.48550/arXiv.2012.15409

[12] Liu J, Li T, Xie P, Du S, Teng F, Yang X, (2020). Urban big data fusion based on deep learning: An overview. Information Fusion, 53: 123–133

[13] Ma X, Fang G, Wang X, (2023). Llm-pruner: On the structural pruning of large language models. Advances in Neural Information Processing Systems, 36: 21702–21720

[14] Nicolas C, Kim J, Chi S, (2021). Natural language processing-based characterization of top-down communication in smart cities for enhancing citizen alignment. Sustainable Cities and Society, 66: 102674

[15] PalmerL (2024). How small but smart cities are changing the way data and decisions are influencing urban life. Nature Cities. doi: 10.1038/s44284-024-00152-3

[16] Tyagi N, Bhushan B, (2023). Demystifying the role of natural language processing (NLP) in smart city applications: Background, motivation, recent advances, and future research directions. Wireless Personal Communications, 130( 2): 857–908

[17] UllahAQiGHussainSUllahIAliZ (2024). The role of llms in sustainable smart cities: Applications, challenges, and future directions. doi: 10.48550/arXiv.2402.14596

[18] Wang J, Hsieh C Y, Wang M, Wang X, Wu Z, Jiang D, Liao B, Zhang X, Yang B, He Q, Cao D, Chen X, Hou T, (2021). Multi-constraint molecular generation based on conditional transformer, knowledge distillation and reinforcement learning. Nature Machine Intelligence, 3( 10): 914–922

[19] WangJJiangRYangCWuZOnizukaMShibasakiRXiaoC (2024). Large language models as urban residents: An LLM agent framework for personal mobility generation. doi: 10.48550/arXiv.2402.14744

[20] Wang S, Zhang X, Ye P, Du M, Lu Y, Xue H, (2019). Geographic knowledge graph (GeoKG): A formalized geographic knowledge representation. ISPRS International Journal of Geo-Information, 8( 4): 184

[21] WeiTLinYLinYGuoSHuJCongGWanH (2024). PTR: A pre-trained language model for trajectory recovery. doi: 10.48550/arXiv.2410.14281

[22] Wong Y Z, Hensher D A, Mulley C, (2020). Mobility as a service (MaaS): Charting a future context. Transportation Research Part A, Policy and Practice, 131: 5–19

[23] Yao Z, Yazdani Aminabadi R, Zhang M, Wu X, Li C, He Y, (2022). Zeroquant: Efficient and affordable post-training quantization for large-scale transformers. Advances in Neural Information Processing Systems, 35: 27168–27183

[24] Zhang K, Zhou F, Wu L, Xie N, He Z, (2024). Semantic understanding and prompt engineering for large-scale traffic data imputation. Information Fusion, 102: 102038

[25] Zhang S, Fu D, Liang W, Zhang Z, Yu B, Cai P, Yao B, (2024a). Trafficgpt: Viewing, processing and interacting with traffic foundation models. Transport Policy, 150: 95–105

[26] ZhangZSunYWangZNieYMaXSunPLiR (2024b). Large language models for mobility in transportation systems: A survey on forecasting tasks. doi: 10.48550/arXiv.2405.02357

[27] ZhengOAbdel-AtyMWangDWangZDingS (2023). ChatGPT is on the horizon: Could a large language model be suitable for intelligent traffic safety research and applications? doi: 10.48550/arXiv.2303.05382

[28] ZhouZLinYJinDLiY (2024a). Large language model for participatory urban planning. doi: 10.48550/arXiv.2402.17161

[29] ZhouZZhouBLiuH (2024b). DynamicRouteGPT: A real-time multi-vehicle dynamic navigation framework based on large language models. doi: 10.48550/arXiv.2408.14185

RIGHTS & PERMISSIONS

Higher Education Press