[1]GUO Shiyuan,WANG Jiayin,SUN Peijie,et al.Collaborative signal enhanced LLM user profiling and recommendation[J].CAAI Transactions on Intelligent Systems,2026,21(2):487-497.[doi:10.11992/tis.202506031]
Copy

Collaborative signal enhanced LLM user profiling and recommendation

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 21 Number of periods: 2026 2 Page number: 487-497 Column: 学术论文—智能系统 Public date: 2026-03-05
Title:
Collaborative signal enhanced LLM user profiling and recommendation
Author(s):
GUO Shiyuan;  WANG Jiayin;  SUN Peijie;  ZHANG Min
Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
Keywords:
information retrieval; recommender system; large language model; user profile; user modeling; contrastive learning; collaborative filtering; user representation; feature embedding
CLC:
TP311.13；TP309
DOI:
10.11992/tis.202506031
Abstract:
The quality of the user profile directly affects the performance of the recommender system. User profile in a traditional recommender system can be derived through modeling the collaborative information between users and items, but is unable to fully utilize the text description information of users and items. The textual information processing and commonsense reasoning capabilities of LLMs, combined with their world knowledge, provide new opportunities for user profiling. The combination of a recommender system and LLM can give full play to the advantages of each other, and improve each other’s performance mutually. This paper proposes a method to introduce two collaborative signals, named potential interest and collaborative scale, from a recommender system into LLM to further enhance the user profile generation of LLM. The user profile is generated through multiple times of interaction with LLM, and is further transformed into an embedding, fusing with the user representation in the recommender system through contrastive learning to improve recommendation performance. Experimental results on two datasets and multiple recommender models show that the proposed method can significantly improve the performance of the recommender model. The proposed method bridges the gap between LLM and recommender system, and sheds light on further similar research work.
References:
[1] CHEN Hao, BEI Yuanchen, SHEN Qijie, et al. Macro graph neural networks for online billion-scale recommender systems[C]//Proceedings of the ACM Web Conference 2024. Singapore: ACM, 2024: 3598-3608.
[2] FENG Zhichao, XIE Junjie, LI Kaiyuan, et al. Context-based fast recommendation strategy for long user behavior sequence in Meituan Waimai[C]//Companion Proceedings of the ACM Web Conference 2024. Singapore: ACM, 2024: 355-363.
[3] DONG Zhikang, LIU Xiulong, CHEN Bin, et al. MuseChat: a conversational music recommendation system for videos[C]//2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle: IEEE, 2024: 12775-12785.
[4] HASAN M R, FERDOUS J. Dominance of AI and machine learning techniques in hybrid movie recommendation system applying text-to-number conversion and cosine similarity approaches[J]. Journal of computer science and technology studies, 2024, 6(1): 94-102
[5] YANG Xinran, ZHANG Liaoniao, FENG Zixin. Personalized tourism recommendations and the E-tourism user experience[J]. Journal of travel research, 2024, 63(5): 1183-1200
[6] BILLSUS D, PAZZANI M J. Learning collaborative information filters[C]//International Conference on Machine Learning. Madison: PMLR, 1998: 46-54.
[7] SARWAR B, KARYPIS G, KONSTAN J, et al. Item-based collaborative filtering recommendation algorithms[C]//Proceedings of the 10th International Conference on World Wide Web. Hong Kong: ACM, 2001: 285-295.
[8] PURIFICATO E, BORATTO L, DE LUCA E W. User modeling and user profiling: a comprehensive survey[EB/OL]. (2024-02-15)[2025-06-25]. https://arxiv.org/abs/2402.09660.
[9] ADOMAVICIUS G, TUZHILIN A. Expert-driven validation of rule-based user models in personalization applications[J]. Data mining and knowledge discovery, 2001, 5(1): 33-58
[10] CHEUNG CHIU B, WEBB G I. Using decision trees for agent modeling: improving prediction performance[J]. User modeling and user-adapted interaction, 1998, 8(1): 131-152
[11] KIM H N, ALKHALDI A, EL SADDIK A, et al. Collaborative user modeling with user-generated tags for social recommender systems[J]. Expert systems with applications, 2011, 38(7): 8488-8496
[12] S?NCHEZ P, BELLOG?N A. Building user profiles based on sequences for content and collaborative filtering[J]. Information processing & management, 2019, 56(1): 192-211
[13] P N, SAITEJA K, RAM K K, et al. University recommender system based on student profile using feature weighted algorithm and KNN[C]//2022 International Conference on Sustainable Computing and Data Communication Systems. Erode: IEEE, 2022: 479-484.
[14] ZIGORIS P, ZHANG Yi. Bayesian adaptive user profiling with explicit & implicit feedback[C]//Proceedings of the 15th ACM International Conference on Information and Knowledge Management. Arlington: ACM, 2006: 397.
[15] ZHANG Yi, KOREN J. Efficient Bayesian hierarchical user modeling for recommendation system[C]//Proceedings of the 30th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. Amsterdam: ACM, 2007: 47-54.
[16] XIE Yi, YU Shunzheng. A large-scale hidden semi-Markov model for anomaly detection on user browsing behaviors[J]. IEEE/ACM transactions on networking, 2009, 17(1): 54-65
[17] TANG Jie, YAO Limin, ZHANG Duo, et al. A combination approach to web user profiling[J]. ACM transactions on knowledge discovery from data, 2010, 5(1): 1-44
[18] AHMAD S B S, RAFIE M, GHORABIE S M. Spam detection on Twitter using a support vector machine and users’ features by identifying their interactions[J]. Multimedia tools and applications, 2021, 80(8): 11583-11605
[19] CHEN Tinggui, YIN Xiaohua, PENG Lijuan, et al. Monitoring and recognizing enterprise public opinion from high-risk users based on user portrait and random forest algorithm[J]. Axioms, 2021, 10(2): 106
[20] TRIPATHI A, T S A, GUDDETI R M R. A reinforcement learning and recurrent neural network based dynamic user modeling system[C]//2018 IEEE 18th International Conference on Advanced Learning Technologies. Mumbai: IEEE, 2018: 411-415.
[21] YU Zeping, LIAN Jianxun, MAHMOODY A, et al. Adaptive user modeling with long and short-term preferences for personalized recommendation[C]//Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. Macao: IJCAI, 2019: 4213-4219.
[22] GE Songwei, DOU Zhicheng, JIANG Zhengbao, et al. Personalizing search results using hierarchical RNN with query-aware attention[C]//Proceedings of the 27th ACM International Conference on Information and Knowledge Management. Torino: ACM, 2018: 347-356.
[23] ZHU Y, LI Hao, LIAO Yikang, et al. What to do next: modeling user behaviors by time-LSTM[C]//International Joint Conference on Artificial Intelligence. Melbourne: IJCAI, 2017: 3602-3608.
[24] MA Zhengyi, DOU Zhicheng, BIAN Guanyue, et al. PSTIE: time information enhanced personalized search[C]//Proceedings of the 29th ACM International Conference on Information & Knowledge Management. Virtual Event: ACM, 2020: 1075-1084.
[25] GU Yulong, DING Zhuoye, WANG Shuaiqiang, et al. Hierarchical user profiling for E-commerce recommender systems[C]//Proceedings of the 13th International Conference on Web Search and Data Mining. Houston: ACM, 2020: 223-231.
[26] QI Tao, WU Fangzhao, WU Chuhan, et al. News recommendation with candidate-aware user modeling[C]//Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. Madrid: ACM, 2022: 1917-1921.
[27] AVNY BROSH T, LIVNE A, SAR SHALOM O, et al. BRUCE: bundle recommendation using contextualized item embeddings[C]//Proceedings of the 16th ACM Conference on Recommender Systems. Seattle: ACM, 2022: 237-245.
[28] CHENG Zhiyong, HAN Sai, LIU Fan, et al. Multi-behavior recommendation with cascading graph convolution networks[C]//Proceedings of the ACM Web Conference 2023. Austin: ACM, 2023: 1181-1189.
[29] HAN Jinkun, LI Wei, CAI Zhipeng, et al. Multi-aggregator time-warping heterogeneous graph neural network for personalized micro-video recommendation[C]//Proceedings of the 31st ACM International Conference on Information & Knowledge Management. Atlanta: ACM, 2022: 676-685.
[30] WANG Xiaolin, SUN Guohao, FANG Xiu, et al. Modeling spatio-temporal neighbourhood for personalized point-of-interest recommendation[C]//Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence. Vienna: IJCAI, 2022: 3530-3536.
[31] TANG Xiangru, ZHANG Xingyao, SHAO Yanjun, et al. Step-back profiling: distilling user history for personalized scientific writing[EB/OL]. (2024-06-20)[2025-06-25]. https://arxiv.org/abs/2406.14275.
[32] ZHANG Jiarui. Guided profile generation improves personalization with LLMs[EB/OL]. (2024-09-19)[2025-06-25]. https://arxiv.org/abs/2409.13093.
[33] SUN Zhongxiang, SI Zihua, ZANG Xiaoxue, et al. Large language models enhanced collaborative filtering[C]//Proceedings of the 33rd ACM International Conference on Information and Knowledge Management. Boise: ACM, 2024: 2178-2188.
[34] ZHOU J, DAI Yijia, JOACHIMS T. Language-based user profiles for recommendation[EB/OL]. (2024-02-23)[2025-06-25]. https://arxiv.org/abs/2402.15623.
[35] JALAN R, PRAKASH T, PEDANEKAR N. LLM-BRec: personalizing session-based social recommendation with LLM-BERT fusion framework[C]//The Second Workshop on Generative Information Retrieval. Washington: ACM, 2024: 1-10.
[36] XI Yunjia, LIU Weiwen, LIN Jianghao, et al. Towards open-world recommendation with knowledge augmentation from large language models[C]//18th ACM Conference on Recommender Systems. Bari: ACM, 2024: 12-22.
[37] WEI Wei, REN Xubin, TANG Jiabin, et al. LLMRec: large language models with graph augmentation for recommendation[C]//Proceedings of the 17th ACM International Conference on Web Search and Data Mining. Merida: ACM, 2024: 806-815.
[38] REN Xubin, WEI Wei, XIA Lianghao, et al. Representation learning with large language models for recommendation[C]//Proceedings of the ACM Web Conference 2024. Singapore: ACM, 2024: 3464-3475.
[39] TANG Zuoli, HUAN Zhaoxin, LI Zihao, et al. One model for all: large language models are domain-agnostic recommendation systems[J]. ACM transactions on information systems, 2025, 43(5): 1-27
[40] DU Yingpeng, WANG Ziyan, SUN Zhu, et al. Large language model with graph convolution for recommendation[EB/OL]. (2024-02-14)[2025-06-25]. https://arxiv.org/abs/2402.08859.
[41] ZHENG Zhi, CHAO Wenshuo, QIU Zhaopeng, et al. Harnessing large language models for text-rich sequential recommendation[C]//Proceedings of the ACM Web Conference 2024. Singapore: ACM, 2024: 3207-3216.
[42] EBRAT D, RUEDA L. Lusifer: LLM-based user simulated feedback environment for online recommender systems[C]//2025 IEEE 4th International Conference on Computing and Machine Intelligence. MI: IEEE, 2025: 1-6.
[43] ZHANG Yang, FENG Fuli, ZHANG Jizhi, et al. CoLLM: integrating collaborative embeddings into large language models for recommendation[J]. IEEE transactions on knowledge and data engineering, 2025, 37(5): 2329-2340
[44] KIM S, KANG H, CHOI S, et al. Large language models meet collaborative filtering: an efficient all-round LLM-based recommender system[C]//Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. Barcelona: ACM, 2024: 1395-1406.
[45] LIU Zhongzhou, ZHANG Hao, DONG Kuicai, et al. Collaborative cross-modal fusion with large language model for recommendation[C]//Proceedings of the 33rd ACM International Conference on Information and Knowledge Management. Boise: ACM, 2024: 1565-1574.
[46] ZHAO Chuang, SU Xing, HE Ming, et al. Collaborative knowledge fusion: a novel method for multi-task recommender systems via LLMs[J]. IEEE transactions on knowledge and data engineering, 2025, 37(9): 5017-5033
[47] VAN DEN OORD A, LI Yazhe, VINYALS O. Representation learning with contrastive predictive coding[EB/OL]. (2018-07-10)[2025-06-25]. https://arxiv.org/abs/1807.03748.
[48] HARPER F M, KONSTAN J A. The MovieLens datasets: history and context[J]. ACM transactions on interactive intelligent systems, 2016, 5(4): 1-19
[49] NI Jianmo, LI Jiacheng, MCAULEY J. Justifying recommendations using distantly-labeled reviews and fine-grained aspects[C]//Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Hong Kong: Association for Computational Linguistics, 2019: 188-197.
[50] J?RVELIN K, KEK?L?INEN J. Cumulated gain-based evaluation of IR techniques[J]. ACM transactions on information systems, 2002, 20(4): 422-446
[51] BUCKLEY C, VOORHEES E M. Retrieval evaluation with incomplete information[C]//Proceedings of the 27th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. Sheffield: ACM, 2004: 25-32.
[52] RENDLE S. Factorization machines[C]//2010 IEEE International Conference on Data Mining. Sydney: IEEE, 2011: 995-1000.
[53] MAO Kelong, ZHU Jieming, WANG Jinpeng, et al. SimpleX: a simple and strong baseline for collaborative filtering[C]//Proceedings of the 30th ACM International Conference on Information & Knowledge Management. Virtual Event: ACM, 2021: 1243-1252.
[54] HE Xiangnan, DENG Kuan, WANG Xiang, et al. LightGCN: simplifying and powering graph convolution network for recommendation[C]//Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval. Virtual Event: ACM, 2020: 639-648.
[55] YU Junliang, XIA Xin, CHEN Tong, et al. XSimGCL: towards extremely simple graph contrastive learning for recommendation[J]. IEEE transactions on knowledge and data engineering, 2024, 36(2): 913-926
[56] CAI Xuheng, HUANG Chao, XIA Lianghao, et al. LightGCL: simple yet effective graph contrastive learning for recommendation[EB/OL]. (2023-02-16)[2025-06-25]. https://arxiv.org/abs/2302.08191.
Similar References:

Memo

-

Last Update: 1900-01-01

Copyright © CAAI Transactions on Intelligent Systems