[1]吴国栋,秦辉,胡全兴,等.大语言模型及其个性化推荐研究[J].智能系统学报,2024,19(6):1351-1365.[doi:10.11992/tis.202309036]
WU Guodong,QIN Hui,HU Quanxing,et al.Research on large language models and personalized recommendation[J].CAAI Transactions on Intelligent Systems,2024,19(6):1351-1365.[doi:10.11992/tis.202309036]
点击复制
WU Guodong,QIN Hui,HU Quanxing,et al.Research on large language models and personalized recommendation[J].CAAI Transactions on Intelligent Systems,2024,19(6):1351-1365.[doi:10.11992/tis.202309036]
大语言模型及其个性化推荐研究
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
19
期数:
2024年第6期
页码:
1351-1365
栏目:
综述
出版日期:
2024-12-05
- Title:
- Research on large language models and personalized recommendation
- Keywords:
- large language model; recommendation; deep learning; supervised fine-tuning; alignment; prompt learning; generative; multimodal
- 分类号:
- TP301
- 摘要:
- 大语言模型因其强大的自然语言处理能力在人工智能领域产生了巨大影响，使得大语言模型个性化推荐成为当前推荐系统研究的新兴领域。本文在深入分析已有大语言模型及其个性化推荐相关研究基础上，探讨大语言模型推荐的过程，并从直接推荐、基于表示学习推荐、基于生成性学习推荐和提示学习推荐四方面详细分析了大语言模型推荐主要的研究进展。指出现有大语言模型推荐研究中存在的推荐偏差、提示脆弱性、有限上下文、高延迟、公平性和评估等问题，展望未来大语言模型推荐研究的主要方向，包括大语言模型推荐的安全性、面向领域的大语言模型推荐、跨模态大语言模型推荐、融合检索任务的大语言模型推荐以及大语言模型推荐的可解释性等。
- Abstract:
- Large language models have revolutionized natural language processing within artificial intelligence, significantly advancing personalized recommendation systems. This paper provides an in-depth analysis of existing research on large language models and their application in personalized recommendations. It explores the process of large language model recommendation and thoroughly analyzes the main research advancements from four perspectives: direct recommendation, representation learning-based recommendation, generation-based recommendation, and prompt learning recommendation. The study identifies several challenges in current research on large language model recommendation, including recommendation bias, vulnerability to prompts, limited contextual understanding, high latency, fairness issues, and evaluation difficulties. It also presents future directions for research on large language model recommendation, including enhancing the security of large language model recommendations, developing domain-oriented large language model recommendations, exploring cross-modal large language model recommendations, integrating retrieval tasks with large language model recommendations, and improving the interpretability of large language model recommendations.
- 参考文献/References:
-
[1] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[EB/OL]. (2017-06-12)[2023-08-25]. http://arxiv.org/abs/1706.03762.
[2] BROWN T B, MANN B, RYDER N, et al. Language models are few-shot learners[EB/OL]. (2020-05-28) [2023-08-25]. http://arxiv.org/abs/2005.14165.
[3] HOU Yupeng, ZHANG Junjie, LIN Zihan, et al. Large language models are zero-shot rankers for recommender systems[EB/OL]. (2023-05-15)[2023-08-25]. http://arxiv.org/abs/2305.08845.
[4] WANG Lei, LIM E P. Zero-shot next-item recommendation using large pretrained language models[EB/OL]. (2023-04-06)[2023-08-25]. http://arxiv.org/abs/2304.03153.
[5] KOJIMA T, GU S S, REID M, et al. Large language models are zero-shot reasoners[EB/OL]. (2022-05-24)[2023-08-25]. http://arxiv.org/abs/2205.11916.
[6] HE Xiangnan, LIAO Lizi, ZHANG Hanwang, et al. Neural collaborative filtering[C]//Proceedings of the 26th International Conference on World Wide Web. [S.l.: s.n.], 2017: 173–182.
[7] 黄璐, 林川杰, 何军, 等. 融合主题模型和协同过滤的多样化移动应用推荐[J]. 软件学报, 2017, 28(3): 708-720.
HUANG Lu, LIN Chuanjie, HE Jun, et al. Diversified mobile app recommendation combining topic model and collaborative filtering[J]. Journal of software, 2017, 28(3): 708-720.
[8] SON J, KIM S B. Content-based filtering for recommendation systems using multiattribute networks[J]. Expert systems with applications, 2017, 89: 404-412.
[9] 刘建勋, 石敏, 周栋, 等. 基于主题模型的Mashup标签推荐方法[J]. 计算机学报, 2017, 40(2): 520-534.
LIU Jianxun, SHI Min, ZHOU Dong, et al. Topic model based tag recommendation method for mashups[J]. Chinese journal of computers, 2017, 40(2): 520-534.
[10] BASILICO J, HOFMANN T. Unifying collaborative and content-based filtering[C]//Twenty-first international conference on Machine learning. Banff: ACM, 2004: 65-72.
[11] 曹俊豪, 李泽河, 江龙, 等. 一种融合协同过滤和用户属性过滤的混合推荐算法[J]. 电子设计工程, 2018, 26(9): 60-63.
CAO Junhao, LI Zehe, JIANG Long, et al. A hybrid recommendation algorithm based on collaborative filtering and user attribute filtering[J]. Electronic design engineering, 2018, 26(9): 60-63.
[12] 孙冬婷, 何涛, 张福海. 推荐系统中的冷启动问题研究综述[J]. 计算机与现代化, 2012(5): 59-63.
SUN Dongting, HE Tao, ZHANG Fuhai. Survey of cold-start problem in collaborative filtering recommender system[J]. Computer and modernization, 2012(5): 59-63.
[13] DA’U A, SALIM N. Recommendation system based on deep learning methods: a systematic review and new directions[J]. Artificial intelligence review, 2020, 53(4): 2709-2748.
[14] ZHANG Jiawei. Graph-ToolFormer: to empower LLMs with graph reasoning ability via prompt augmented by ChatGPT[EB/OL]. (2023-04-10)[2023-08-25]. http://arxiv.org/abs/2304.11116.
[15] PAPARRIZOS I, CAMBAZOGLU B B, GIONIS A. Machine learned job recommendation[C]//Proceedings of the fifth ACM conference on Recommender systems. Chicago: ACM, 2011: 325-328.
[16] LIU Jiahui, DOLAN P, PEDERSEN E R. Personalized news recommendation based on click behavior[C]//Proceedings of the 15th international conference on Intelligent user interfaces. Hong Kong: ACM, 2010: 31-40.
[17] DAI Sunhao, SHAO Ninglu, ZHAO Haiyuan, et al. Uncovering ChatGPT’s capabilities in recommender systems[EB/OL]. (2023-05-03)[2023-08-25]. http://arxiv.org/abs/2305.02182.
[18] WANG Lei, ZHANG Jingsen, CHEN Xu, et al. User behavior simulation with large language model based agents[EB/OL]. (2023-06-05)[2023-08-25]. https://arxiv.org/abs/2306.02552.
[19] 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.
[20] 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.
[21] COLIN R, NOAM S, ADAM R, et al. Exploring the limits of transfer learning with a unified text-to-text transformer[J]. Journal of machine learning research, 2020, 21(1): 5485-5551.
[22] LI Jiacheng, WANG Ming, LI Jin, et al. Text is all you need: learning language representations for sequential recommendation[EB/OL]. (2023-05-23)[2023-08-25]. http://arxiv.org/abs/2305.13731.
[23] CUI Zeyu, MA Jianxin, ZHOU Chang, et al. M6-rec: generative pretrained language models are open-ended recommender systems[EB/OL]. (2022-05-17) [2023-08-25]. http://arxiv.org/abs/2205.08084.
[24] KOREN Y, BELL R, VOLINSKY C. Matrix factorization techniques for recommender systems[J]. Computer, 2009, 42(8): 30-37.
[25] 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.
[26] FRIEDMAN L, AHUJA S, ALLEN D, et al. Leveraging large language models in conversational recommender systems[EB/OL]. (2023-05-13)[2023-08-25]. http://arxiv.org/abs/2305.07961.
[27] LIN Guo, ZHANG Yongfeng. Sparks of artificial general recommender (AGR): early experiments with ChatGPT[EB/OL]. (2017-05-08)[2023-08-25]. http://arxiv.org/abs/2305.04518.
[28] XI Yunjia, LIU Weiwen, LIN Jianghao, et al. Towards open-world recommendation with knowledge augmentation from large language models[EB/OL]. (2023-06-19) [2023-08-25]. http://arxiv.org/abs/2306.10933.
[29] LI Jinming, ZHANG Wentao, WANG Tian, et al. GPT4Rec: a generative framework for personalized recommendation and user interests interpretation[EB/OL]. (2023-04-08)[2023-04-25]. http://arxiv.org/abs/2304.03879.
[30] DU Yingpeng, LUO Di, YAN Rui, et al. Enhancing job recommendation through LLM-based generative adversarial networks[EB/OL]. (2023-07-20)[2023-08-25]. http://arxiv.org/abs/2307.10747.
[31] WANG Wenjie, LIN Xinyu, FENG Fuli, et al. Generative recommendation: towards next-generation recommender paradigm[EB/OL]. (2023-04-07)[2023-08-25]. http://arxiv.org/abs/2304.03516.
[32] GENG Shijie, LIU Shuchang, FU Zuohui, et al. Recommendation as language processing (RLP): a unified pretrain, personalized prompt & predict paradigm (P5)[C]//Proceedings of the 16th ACM Conference on Recommender Systems. Seattle: ACM, 2022: 299-315.
[33] JI Jianchao, LI Zelong, XU Shuyuan, et al. GenRec: large language model for Generative recommendation[M]//Lecture Notes in Computer Science. Cham: Springer Nature Switzerland, 2024: 494-502.
[34] ZHANG Zizhuo, WANG Bang. Prompt learning for news recommendation[EB/OL]. (2023-04-11)[2023-08-25]. http://arxiv.org/abs/2304.05263.
[35] OKAMOTO K, CHEN Wei, LI Xiangyang. Ranking of closeness centrality for large-scale social networks[C]//PREPARATA FP, WU X, YIN J. International Workshop on Frontiers in Algorithmics. Berlin: Springer, 2008: 186-195.
[36] ZHANG Junlong, LUO Yu. Degree centrality, betweenness centrality, and closeness centrality in social network[C]//Proceedings of the 2017 2nd International Conference on Modelling, Simulation and Applied Mathematics. Paris: Atlantis Press, 2017: 300-303.
[37] NEWMAN M E J. A measure of betweenness centrality based on random walks[J]. Social networks, 2005, 27(1): 39-54.
[38] HUANG Xiao, ZHANG Jingyuan, LI Dingcheng, et al. Knowledge graph embedding based question answering[C]//Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining. Melbourne: ACM, 2019: 105-113.
[39] ZHANG Yuyu, DAI Hanjun, KOZAREVA Z, et al. Variational reasoning for question answering with knowledge graph[EB/OL]. (2017-11-27)[2023-01-01]. http://arxiv.org/abs/1709.04071.
[40] RONG Yu, HUANG Wenbing, XU Tingyang, et al. DropEdge: towards deep graph convolutional networks on node classification[EB/OL]. (2019-07-25)[2023-08-25]. http://arxiv.org/abs/1907.10903.
[41] ERRICA F, PODDA M, BACCIU D, et al. A fair comparison of graph neural networks for graph classification[EB/OL]. (2019-12-20)[2023-08-25]. http://arxiv.org/abs/1912.09893.
[42] WANG Heng, FENG Shangbin, HE Tianxing, et al. Can language models solve graph problems in natural language?[EB/OL]. (2023-05-17)[2023-08-25]. http://arxiv.org/abs/2305.10037.
[43] XIE Han, ZHENG Da, MA Jun, et al. Graph-aware language model pre-training on a large graph corpus can help multiple graph applications [EB/OL]. (2023-06-05) [2023-08-25]. http://arxiv.org/abs/2306.02592.
[44] GUO Jiayan, DU Lun, LIU Hengyu, et al. GPT4Graph: can large language models understand graph structured data? an empirical evaluation and benchmarking [EB/OL]. (2023-05-24) [2023-08-25]. http://arxiv.org/abs/2305.15066.
[45] WU Likang, QIU Zhaopeng, ZHENG Zhi, et al. Exploring large language model for graph data understanding in online job recommendations[EB/OL]. (2023-07-10)[2023-08-25]. http://arxiv.org/abs/2307.05722.
[46] ZHU Deyao, CHEN Jun, SHEN Xiaoqian, et al. MiniGPT-4: enhancing vision-language understanding with advanced large language models [EB/OL]. (2023-04-20) [2023-08-25]. http://arxiv.org/abs/2304.10592.
[47] LI Junnan, LI Dongxu, SAVARESE S, et al. BLIP-2: bootstrapping language-image pre-training with frozen image encoders and large language models [EB/OL]. (2023-01-30) [2023-08-25]. http://arxiv.org/abs/2301.12597.
[48] BAO Keqin, ZHANG Jizhi, ZHANG Yang, et al. TALLRec: an effective and efficient tuning framework to align large language model with recommendation [EB/OL]. (2023-04-30) [2023-08-25]. http://arxiv.org/abs/2305.00447.
[49] ZHANG Junjie, XIE Ruobing, HOU Yupeng, et al. Recommendation as instruction following: a large language model empowered recommendation approach [EB/OL]. (2023-05-11) [2023-08-25]. http://arxiv.org/abs/2305.07001.
[50] KANG Wangcheng, NI Jianmo, MEHTA N, et al. Do LLMs understand user preferences? evaluating LLMs on user rating prediction [EB/OL]. (2023-05-10) [2023-08-25]. http://arxiv.org/abs/2305.06474.
[51] YONA G, HONOVICH O, LAISH I, et al. Surfacing biases in large language models using contrastive input decoding[EB/OL]. (2023-05-12)[2023-08-25]. http://arxiv.org/abs/2305.07378.
[52] CHEN Zhipeng, ZHOU Kun, ZHANG Beichen, et al. ChatCoT: tool-augmented chain-of-thought reasoning on chat-based large language models[EB/OL]. (2023-05-23) [2023-08-25]. http://arxiv.org/abs/2305.14323.
[53] OUYANG Long, WU J, XU Jiang, et al. Training language models to follow instructions with human feedback[EB/OL]. (2022-03-04)[2023-08-25]. http://arxiv.org/abs/2203.02155.
[54] HUA Wenyue, GE Yingqiang, XU Shuyuan, et al. UP5: unbiased foundation model for fairness-aware recommendation[EB/OL]. (2023-05-20)[2023-08-25]. http://arxiv.org/abs/2305.12090.
[55] ZHANG Jizhi, BAO Keqin, ZHANG Yang, et al. Is ChatGPT fair for recommendation? evaluating fairness in large language model recommendation[EB/OL]. (2023-05-12) [2023-08-25]. http://arxiv.org/abs/2305.07609.
[56] AI Qingyao, BAI Ting, CAO Zhao, et al. Information retrieval meets large language models: a strategic report from chinese IR community[J]. AI open, 2023, 4: 80-90.
[57] ZHAO W X, ZHOU Kun, LI Junyi, et al. A survey of large language models[EB/OL]. (2023-03-21)[2023-08-25]. http://arxiv.org/abs/2303.18223.
[58] BILLS S, CAMMARATA N, MOSSING D, et al. Language models can explain neurons in language models[EB/OL]. (2023-05-09)[2024-03-25]. https://openaipublic.blob. core. windows. net/neuron-explainer/paper/index.
[59] WEI Wei, REN Xubin, TANG Jiabin, et al. LLMRec: large language models with graph augmentation for recommendation[EB/OL]. (2023-11-01)[2024-03-25]. http://arxiv.org/abs/2311.00423.
- 相似文献/References:
-
[1]李慧,马小平,胡云,等.融合上下文信息的社会网络推荐系统[J].智能系统学报,2015,10(2):293.[doi:10.3969/j.issn.1673-4785.201406017]
LI Hui,MA Xiaoping,HU Yun,et al.Social network recommendaton system mixing contex information[J].CAAI Transactions on Intelligent Systems,2015,10():293.[doi:10.3969/j.issn.1673-4785.201406017]
[2]王大玲,冯时,张一飞,等.社会媒体多模态、多层次资源推荐技术研究[J].智能系统学报,2014,9(3):265.[doi:10.3969/j.issn.1673-4785.201403068]
WANG Daling,FENG Shi,ZHANG Yifei,et al.Study on the recommendations of multi-modal and multi-level resources in social media[J].CAAI Transactions on Intelligent Systems,2014,9():265.[doi:10.3969/j.issn.1673-4785.201403068]
[3]顾军华,谢志坚,武君艳,等.基于图游走的并行协同过滤推荐算法[J].智能系统学报,2019,14(4):743.[doi:10.11992/tis.201806002]
GU Junhua,XIE Zhijian,WU Junyan,et al.Parallel collaborative filtering recommendation algorithm based on graph walk[J].CAAI Transactions on Intelligent Systems,2019,14():743.[doi:10.11992/tis.201806002]
[4]黄河燕,李思霖,兰天伟,等.大语言模型安全性：分类、评估、归因、缓解、展望[J].智能系统学报,2025,20(1):2.[doi:10.11992/tis.202401006]
HUANG Heyan,LI Silin,LAN Tianwei,et al.A survey on the safety of large language model: classification, evaluation, attribution, mitigation and prospect[J].CAAI Transactions on Intelligent Systems,2025,20():2.[doi:10.11992/tis.202401006]
备注/Memo
收稿日期:2023-9-21。
基金项目:国家自然科学基金项目(32371993)；安徽省自然科学基金项目(2108085MF209)；安徽省科技重大专项(202103b06020013).
作者简介:吴国栋，副教授，主要研究方向为深度学习、推荐系统。主持安徽省自然科学研究重点项目 1 项、一般项目 1 项、安徽省科技攻关重点项目 1 项。发表学术论文30 余篇。E-mail：wugd@ahau.edu.cn;秦辉，硕士研究生，主要研究方向为推荐系统。E-mail：2504864202@qq.com;胡全兴，硕士研究生，主要研究方向区块链可信推荐。E-mail：1763273299@qq.ocm。
通讯作者:吴国栋. E-mail：gdwu1120@qq.com
更新日期/Last Update:
2024-11-05