[1]XIE Guangming,BAI Yanbing,WU Ziang,et al.Review of LLM-based recommendation systems[J].CAAI Transactions on Intelligent Systems,2025,20(6):1520-1533.[doi:10.11992/tis.202410007]
Copy

Review of LLM-based recommendation systems

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 20 Number of periods: 2025 6 Page number: 1520-1533 Column: 吴文俊人工智能科学技术奖论坛 Public date: 2025-11-05
Title:
Review of LLM-based recommendation systems
Author(s):
XIE Guangming1;  BAI Yanbing1;  WU Ziang2;  ZHANG Yanling2
1. College of Engineering, Peking University, Beijing 100871, China;
2. School of Intelligence Science and Technology, University of Science and Technology Bejing, Beijing 100083, China
Keywords:
recommendation system; recommendation models; LLM; prompt engineering; scaling law; visual large model; sequential modeling; deep learning
CLC:
TP391.3
DOI:
10.11992/tis.202410007
Abstract:
As social networking and e-commerce platforms have grown in popularity, industrial recommendation systems have assumed an increasingly significant role in the mobile internet era. The recommendation system is imperative for enhancing user experience, optimizing shopping experience, and promoting user growth. In the domain of recommendation systems, the role of models is paramount. As computing power and data volume have increased, model structures have become increasingly complex. These models have also improved the accuracy of recommendation systems in comparison to traditional models. Represented by GPT and DeepSeek, LLM has been demonstrated to enhance the efficacy of language models and catalyze the evolution of novel model training paradigms, such as prompt engineering. The rapid advancements in large language model (LLM) capabilities, particularly in semantic understanding and content generation, are poised to transform industrial recommendation systems. This paper reviews the connections between LLM and recommendation systems, then outlines the ways in which LLM can be integrated with industrial recommendation systems. The objective of our work is to leverage technologies associated with LLM to enhance the efficiency and efficacy of recommendation models.
References:
[1] 项亮. 推荐系统实践[M]. 北京: 人民邮电出版社, 2012.
[2] COVINGTON P, ADAMS J, SARGIN E. Deep neural networks for YouTube recommendations[C]//Proceedings of the 10th ACM Conference on Recommender Systems. Boston: ACM, 2016: 191-198.
[3] AIVAZOGLOU M, ROUSSOS A O, MARGARIS D, et al. A fine-grained social network recommender system[J]. Social network analysis and mining, 2019, 10(1): 8.
[4] QUESTMOBILE. QuestMobile2023中国移动互联网年度报告[EB/OL]. (2024-01-30)[2025-02-24]. https://www.questmobile.com.cn/research/report.
[5] GU Siyu, SHENG Xiangrong. On ranking consistency of pre-ranking stage[EB/OL]. (2022-05-03)[2025-02-24]. https://arxiv.org/abs/2205.01289.
[6] ZHAO W X, ZHOU Kun, LI Junyi, et al. A survey oflarge language models[EB/OL]. (2023-05-31)[2025-02-24]. https://arxiv.org/abs/2303.18223.
[7] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.
[8] ELMAN J. Finding structure in time[J]. Cognitive science, 1990, 14(2): 179-211.
[9] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems30 (NIPS2017). Long Beach: Curran Associates Inc, 2017: 5998-6008.
[10] PETERS M, NEUMANN M, IYYER M, et al. Deep contextualized word representations[C]//Proceedings of the 2018 Conference of the North American Chapter Of the Association for Computational Linguistics: Human LanguageTechnologies. New Orleans: USAACL, 2018: 2227-2237.
[11] DEVLIN J, CHANG Mingwei, LEE K, et al. BERT: pre-training of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. Minnesota: Association for Computational Linguistics, 2019: 4171-4186.
[12] OUYANG L, WU J, JIANG Xu, et al. Training language models to follow instructions with human feedback[C]//Advances in Neural Information Processing Systems 35 (NeurIPS 2022). Louisiana: Curran Associates Inc, 2022: 27730-27744.
[13] LEWIS M, LIU Yinhan, GOYAL N, et al. Bart: denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension[EB/OL]. (2019-10-29)[2025-02-24]. https://arxiv.org/abs/1910.13461.
[14] RAFFEL C, SHAZEER N, ROBERTS A, et al. Exploring the limits of transfer learning with aunified text-to-text Transformer[EB/OL]. (2019-10-23)[2025-02-24]. https://artxiv.org/abs/1910.10683.
[15] ZHANG Zhengyan, HAN Xu, LIU Zhiyuan, et al. ERNIE: enhanced language representation with informative entities[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Florence, Stroudsburg: USAACL, 2019: 1441-1451.
[16] LIU Yinhan, OTT M, GOYAL N, et al. RoBERTa: A robustly optimized BERT pretraining approach[EB/OL]. (2019-07-26)[2024-05-21]. https://arxiv.org/abs/1907.11692.
[17] LAN Z, CHEN M, GOODMAN S, et al. Albe-rt: a lite BERT for self-supervised learning of language representations[C]//8th International Conference on Learning Representations. [S.l.]: Open Access, 2020.
[18] LIU Aixin, FENG Bei, XUE Bing, et al. Deepseek-v3 technical report[EB/OL]. (2024-12-27)[2025-01-01]. https://arxiv.org/abs/2412.19437.
[19] GUO D, YANG Dejian, ZHANG Haowei, et al. DeepSeek-R1：incentivizing reasoning capability in LLMs via reinforcement learning[EB/OL]. (2025-01-22)[2025-02-02]. https://arxiv.org/abs/2501.12948.
[20] ACHIAM J, ADLER S, AGARWAL S, et al. GPT-4 technical report[EB/OL]. (2024-05-04)[2025-02-24]. https://arxiv.org/pdf/2303.08774.
[21] ZHANG Susan, ROLLER S, GOYAL N, et al. OPT：open pre-trained Transformer language models[EB/OL]. (2022-05-02)[2025-02-24]. https://arxiv.org/abs/2205.01068.
[22] TOUVRON H, LAVRIL T, IZACARD G, et al. LLaMA: open and efficient foundation language models[EB/OL]. (2023-02-27)[2025-02-24]. https://arxiv.org/abs/2302.13971.
[23] TOUVRON H, MARTIN L, STONE K, et al. LLaMA 2: open foundation and fine-tuned chat models[EB/OL]. (2023-07-18)[2025-02-24]. https://arxiv.org/abs/2307.09288.
[24] DUBEY A, JAUHRIi A, PANDEY A, et al. The LLaMA 3 herd of models[EB/OL]. (2024-07-31)[2025-02-24]. https://arxiv.org/abs/2407.21783.
[25] BAI Yuntao, JONES A, NDOUSSE K, et al. Training a helpful and harmless assistant with reinforcement learning from human feedback[EB/OL]. (2022-04-12)[2025-02-24]. https://arxiv.org/abs/2204.05862.
[26] LIU H, LI C, WU Q, et al. Visual instruction tuning[C]//Advances in Neural Information Processing Systems 36 (NeurIPS2023). Louisiana: Curran Associates Inc, 2023: 34892-34916.
[27] RADFORD A, KIM J W, HALLACY C, et al. Learning transferable visual models from natural language supervision[C]//The 38thInternationalConference on Machine Learning. [S.l.]: Curran Associates Inc, 2021: 8748-8763.
[28] DEY R, SALEM F M. Gate-variants of gated recurrent unit (GRU) neural networks[C]//2017 IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS). Boston: IEEE, 2017: 1597-1600.
[29] LIU Pengfei, YUAN Weizhe, FU Jinlan, et al. Pre-train, prompt, and predict: a systematic survey of prompting methods in natural language processing[J]. ACM computing surveys, 2023, 55(9): 1-35.
[30] WANG Qi, LI Jindong, WANG Shiqi, et al. Towards next-generation LLM-based recommender systems: a survey and beyond[EB/OL]. (2024-10-10)[2025-02-24]. https://arxiv.org/abs/2410.19744.
[31] LIN Jianghao, DAI Xinyi, XI Yunjia, et al. How can recommender systems benefit from large language models: a survey[EB/OL]. (2023-07-09)[2025-02-24]. https://arxiv.org/abs/2306.05817.
[32] ZHAO Zihuai, FAN Wenqi, LI Jiatong, et al. Recommender systems in the era of large language models (LLMS)[EB/OL]. (2023-07-05)[2025-02-24]. https://arxiv.org/abs/2307.02046.
[33] GUO Huifeng, TANG Ruiming, YE Yunming, et al. DeepFM: a factorization-machine based neural network for CTR prediction[EB/OL]. (2017-05-13)[2025-02-24]. https://arxiv.org/abs/1703.04247.
[34] LIAN Jianxun, ZHOU Xiaohuan, ZHANG Fuzheng, et al. xDeepFM: combining explicit and implicit feature interactions for recommender systems[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London: ACM, 2018: 1754-1763.
[35] WANG R, FU B, FU G, et al. Deep and crossnetwork for adclick predictions[C]//Proceedings of the ADKDD’17. Halifax: Associationfor Computing Machinery, 2017: 1-7.
[36] WANG Ruoxi, SHIVANNA R, CHENG D, et al. DCN V2: improved deep & cross network and practical lessons for web-scale learning to rank systems[C]//Proceedings of the Web Conference 2021. Ljubljana: ACM, 2021: 1785-1797.
[37] MA Jiaqi, ZHAO Zhe, YI Xinyang, et al. Modeling task relationships in multi-task learning with multi-gate mixture-of-experts[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London: ACM, 2018: 1930-1939.
[38] TANG Hongyan, LIU Junning, ZHAO Ming, et al. Progressive layered extraction (PLE): a novel multi-task learning (MTL) model for personalized recommendations[C]//Fourteenth ACM Conference on Recommender Systems. Brazil: ACM, 2020: 269-278.
[39] CHANG Jianxin, ZHANG Chenbin, HUI Yiqun, et al. PEPNet: parameter and embedding personalized network for infusing with personalized prior information[C]//Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Long Beach: ACM, 2023: 3795-3804.
[40] SWIETOJANSKI P, LI Jinyu, RENALS S. Learning hidden unit contributions for unsupervised acoustic model adaptation[J]. IEEE/ACM transactions on audio, speech, and language processing, 2016, 24(8): 1450-1463.
[41] ZHOU Guorui, ZHU Xiaoqiang, SONG Chenru, et al. Deep interest network for click-through rate prediction[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London: ACM, 2018: 1059-1068.
[42] ZHOU Guorui, MOU Na, FAN Ying, et al. Deep interest evolution network for click-through rate prediction[J]. Proceedings of the AAAI conference on artificial intelligence, 2019, 33(1): 5941-5948.
[43] PI Qi, ZHOU Guorui, ZHANG Yujing, et al. Search-based user interest modeling with lifelong sequential behavior data for click-through rate prediction[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management. [S.l.]: ACM, 2020: 2685-2692.
[44] CHANG Jianxin, ZHANG Chenbin, FU Zhiyi, et al. TWIN: two-stage interest network for lifelong user behavior modeling in CTR prediction at Kuaishou[C]//Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Long Beach: ACM, 2023: 3785-3794.
[45] ZHU Han, LI Xiang, ZHANG Pengye, et al. Learning tree-based deep model for recommender systems[C]//Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. London: ACM, 2018: 1079-1088.
[46] GAO Weihao, FAN Xiangjun, WANG Chong, et al. Deep retrieval: Learning a retrievable structure for large-scale recommendations[EB/OL]. (2020-07-12)[2025-02-24]. https://arxiv.org/abs/2007.07203.
[47] VAN DEN OORD A, VINYALS O, KAVUKCUOGLU K, et al. Neural discrete representation learning[C]//Advances in Neural Information Processing Systems 30. California: Curran Associates Inc, 2017.
[48] RAJPUT S, MEHTA N, SINGH A, et al. Recommender Systems with Generative Retrieval[C]//Advances in Neural Information Processing Systems 36. Louisiana: Curran Associates Inc, 2023: 10299-10315.
[49] QIU Zhaopeng, WU Xian, GAO Jingyue, et al. U-BERT: pre-training user representations for improved recommendation[J]. Proceedings of the AAAI conference on artificial intelligence, 2021, 35(5): 4320-4327.
[50] DING Hao, MA Yifei, DEORAS A, et al. Zero-shot recommender systems[EB/OL]. (2021-05-18)[2025-02-24]. https://arxiv.org/abs/2105.08318.
[51] HOU Yupeng, MU Shanlei, ZHAO W X, et al. Towards universal sequence representation learning for recommender systems[C]//Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Washington DC: ACM, 2022: 585-593.
[52] HOU Yupeng, HE Zhankui, MCAULEY J, et al. Learning vector-quantized item representation for transferable sequential recommenders[C]//Proceedings of the ACM Web Conference 2023. Austin: ACM, 2023: 1162-1171.
[53] SHAZEER N, MIRHOSEINI A, MAZIARZ K, et al. Outrageously large neural networks: the sparsely-gated mixture-of-experts layer[C]//International Conference on Learning Represent-ations. Toulon: openreview. net, 2017.
[54] GE Tiezheng, HE Kaiming, KE Qifa, et al. Optimized product quantization[J]. IEEE transactions on pattern analysis and machine intelligence, 2014, 36(4): 744-755.
[55] SUN Fei, LIU Jun, WU Jian, et al. BERT4Rec: sequential recommendation with bidirectional encoder representations from transformer[C]//Proceedings of the 28th ACM International Conference on Information and Knowledge Management. Beijing: ACM, 2019: 1441-1450.
[56] WU Chuhan, WU Fangzhao, QI Tao, et al. PTUM: pre-traininguser model fromunlabeled user behaviors via self-supervision[EB/OL]. (2020-10-04)[2025-02-24]. https://arxiv.org/abs/2010.01494.
[57] LIN Junyang, MEN Rui, YANG An, et al. M6: a Chinese multimodal pre-trainer[EB/OL]. (2021-05-01)[2025-02-24]. https://arxiv.org/abs/2103.00823.
[58] WANG Yuhao, ZHAO Xiangyu, CHEN Bo, et al. PLATE: a prompt-enhanced paradigm for multi-scenario recommendations[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. Taipei: ACM, 2023: 1498-1507.
[59] GAO Yunfan, SHENG Tao, XIANG Youlin, et al. Chat-rec: Towards interactive and explainable LLMS-augmented recommender system[EB/OL]. (2023-05-25)[2025-02-24]. https://arxiv.org/abs/2303.14524.
[60] FRIEDMAN L, AHUJA S, ALLEN D, et al. Leveraging large language models in conversational recommender systems[EB/OL]. (2023-05-13)[2025-02-24]. https://arxiv.org/abs/2305.07961.
[61] ZHANG Zizhuo, WANG Bang. Prompt learning for news recommendation[C]//Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. Taipei: ACM, 2023: 227-237.
[62] LIU Yuqing, WANG Yu, SUN Lichao, et al. Rec-GPT4V: multimodal recommendation with large vision-language models[EB/OL]. (2024-02-13)[2025-02-24]. https://arxiv.org/abs/2402.08670.
[63] YE Yuyang, ZHENG Zhi, SHEN Yishan, et al. Harnessing multimodal large language models for multimodal sequential recommendation[EB/OL]. (2024-08-19)[2025-05-20]. https://arxiv.org/abs/2408.09698.
[64] TIAN Jiahao, ZHAO Jinman, WANG Zhenkai, et al. MMREC: LLM based multi-modal recommender system[EB/OL]. (2024-08-08)[2025-02-24]. https://arxiv.org/abs/2408.04211.
[65] ZHAI Jiaqi, LIAO L, LIU Xing, et al. Actions speak louder than words: trillion-parameter sequential transducers for generative recommendations[EB/OL]. (2024-02-27)[2025-02-24]. https://arxiv.org/abs/2402.17152.
[66] ZHANG Buyun, LUO Liang, CHEN Yuxin, et al. Wukong: towards a scaling law for large-scale recom-mendation[EB/OL]. (2024-05-04)[2025-02-24]. https://arxiv.org/abs/2403.02545.
[67] ZHANG Chao, WU Shiwei, ZHANG Haoxin, et al. NoteLLM: a retrievable large language model for note recommendation[C]//Companion Proceedings of the ACM Web Conference 2024. Singapore: ACM, 2024: 170-179.
[68] ZHAO Qian, QIAN Hao, LIU Ziqi, et al. Breaking the barrier: utilizing large language models for industrial recommendation systems through an inferential knowledge graph[C]//Proceedings of the 33rd ACM International Conference on Information and Knowledge Management. Boise: ACM, 2024: 5086-5093.
[69] JIA Jian, WANG Yipei, LI Yan, et al. LEARN: knowledge adaptation from large language model to recommendation for practical industrial application[EB/OL]. (2024-12-26)[2025-02-24]. https://arxiv.org/abs/2405.03988.
[70] CUI Yu, LIU Feng, WANG Pengbo, et al. Distillation matters: empowering sequential recommenders to match the performance of large language models[C]//18th ACM Conference on Recommender Systems. Bari: ACM, 2024: 507-517.
[71] FAN Xinyan, LIU Zheng, LIAN Jianxun, et al. Lighter and better: low-rank decomposed self-attention networks for next-item recommendation[C]//Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval. [S.l.]: ACM, 2021: 1733-1737.
[72] WHITE C, SAFARI M, SUKTHANKER R, et al. Neural architecture search: Insights from 1000 papers[EB/OL]. (2023-01-20)[2025-02-24]. https://arxiv.org/abs/2301.08727.
[73] LI Qinbin, WEN Zeyi, WU Zhaomin, et al. A survey on federated learning systems: vision, hype and reality for data privacy and protection[J]. IEEE transactions on knowledge and data engineering, 2023, 35(4): 3347-3366.
[74] ZHAO Haiyan, CHEN Hanjie, YANG Fan, et al. Explainability for large language models: a survey[J]. ACM transactions on intelligent systems and technology, 2024, 15(2): 1-38.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems