Research on Building a Vertical AI Foundation for Automotive Digital Marketing with Open-Source Large Models

Rui Shi; Haonan Chen

doi:10.62051/ijcsit.v8n6.05

Authors

Rui Shi CATARC Brand Technology (Tianjin) Co., Ltd., China
Haonan Chen CATARC Brand Technology (Tianjin) Co., Ltd., China

DOI:

https://doi.org/10.62051/ijcsit.v8n6.05

Keywords:

Large language model, Automotive digital marketing, Vertical model, Supervised fine-tuning, Retrieval-augmented generation

Abstract

As the automotive industry transitions from "incremental expansion" to "deepening existing markets," digital marketing is emerging as a critical battleground for cost reduction and efficiency enhancement. This study focuses on the "platform AI large model foundation" system, exploring how to integrate open-source large models with domain-specific automotive datasets to train vertical models tailored for digital marketing scenarios in the automotive sector, thereby supporting core capabilities such as semantic understanding and content generation. The research constructs a specialized automotive marketing corpus covering vehicle analysis, competitive benchmarking, user profiling, and marketing messaging. It proposes a dual-model training architecture combining LoRA-based parameter-efficient fine-tuning and RAG (Retrieval-Augmented Generation), and integrates knowledge graphs into the marketing decision-making pipeline. This work provides a practical technical foundation and systematic methodology for the intelligent transformation of digital marketing in the automotive industry.

Downloads

Download data is not yet available.

References

[1] DeepZero. (n.d.). Deep Agent. https://www.deepzero.com/product/alphadata/deepagent/

[2] Sohu. (2025, March 7). KuKa Technology announces strategic transformation: “KuaiChao” integrates DeepSeek for upgraded automotive marketing intelligence. https://www.sohu.com/a/868122895_120389468

[3] Ma, W., Li, Y., & Gao, J. (2026). Research on robot target recognition based on multimodal visual self attention mechanism. Computer Measurement & Control, 34(7), 261 267. https://doi.org/10.16526/j.cnki.11 4762/tp.2026.07.032

[4] Jia, S. (2026). Chinese character relationship extraction based on BiLSTM CNN and attention mechanism. Research and Application of Digital Intelligence Technology. Advance online publication. https://doi.org/10.26917/j.cnki.issn.2097 597X.2026.02.001

[5] Guo, X., Li, B., Wu, Y., Liu, J., & Xu, M. (2026). A visual generation system for ammunition transfer and supply schemes based on semantic graph retrieval enhancement. Journal of Computer Aided Design & Graphics. Advance online publication. https://link.cnki.net/urlid/11.2925.tp.20260806.1649.002

[6] Chen, L., Lei, K., & Liu, C. (2026). Research on question answering method in power grid domain based on graph retrieval augmented generation. Computer Applications and Software. Advance online publication. https://link.cnki.net/urlid/31.1260.TP.20260806.1406.002

[7] Guan, S., Xi, J., & Song, H. (2026). A bidirectional LSTM sentiment analysis method for reviews integrating BERT and attention mechanism. Software Engineering, 29(8), 19 22+28. https://doi.org/10.19644/j.cnki.issn2096 1472.2026.008.004

[8] Zhu, W., Su, B., & Li, H. (2026). A multi classification method for civil aviation news based on BERT and BiGRU. Computer and Modernization, (7), 91 96.

[9] Dou, J., Wang, X., Qiao, J., et al. (2026). An intelligent decision support framework for distribution network maintenance based on knowledge graph retrieval augmented generation and large language models. Engineering Science and Technology, 58(4), 67 79.

[10] Liu, C. (2026). Research on key technologies for retrieval augmented generation of geographical knowledge graphs based on multimodal data [Master’s thesis, Chongqing Jiaotong University]. https://doi.org/10.27671/d.cnki.gcjtc.2026.000508