{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,11]],"date-time":"2025-09-11T18:09:08Z","timestamp":1757614148568,"version":"3.44.0"},"reference-count":36,"publisher":"Association for Computing Machinery (ACM)","issue":"11","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Proc. VLDB Endow."],"published-print":{"date-parts":[[2025,7]]},"abstract":"<jats:p>\n            Graph Neural Networks (GNNs) have demonstrated remarkable effectiveness in recommendation systems. However, conventional graph-based recommenders, such as LightGCN, require maintaining embeddings of size\n            <jats:italic toggle=\"yes\">d<\/jats:italic>\n            for each node, resulting in a parameter complexity of\n            <jats:italic toggle=\"yes\">O<\/jats:italic>\n            (\n            <jats:italic toggle=\"yes\">n<\/jats:italic>\n            X\n            <jats:italic toggle=\"yes\">d<\/jats:italic>\n            ), where\n            <jats:italic toggle=\"yes\">n<\/jats:italic>\n            represents the total number of users and items. This scaling pattern poses significant challenges for deployment on large-scale graphs encountered in real-world applications. To address this scalability limitation, we propose\n            <jats:bold>Lighter-X<\/jats:bold>\n            , an efficient and modular framework that can be seamlessly integrated with existing GNN-based recommender architectures. Our approach substantially reduces both parameter size and computational complexity while preserving the theoretical guarantees and empirical performance of the base models, thereby enabling practical deployment at scale. Specifically, we analyze the original structure and inherent redundancy in their parameters, identifying opportunities for optimization. Based on this insight, we propose an efficient compression scheme for the sparse adjacency structure and high-dimensional embedding matrices, achieving a parameter complexity of\n            <jats:italic toggle=\"yes\">O<\/jats:italic>\n            (\n            <jats:italic toggle=\"yes\">h<\/jats:italic>\n            X\n            <jats:italic toggle=\"yes\">d<\/jats:italic>\n            ), where\n            <jats:italic toggle=\"yes\">h<\/jats:italic>\n            &gt;&gt;\n            <jats:italic toggle=\"yes\">n.<\/jats:italic>\n            Furthermore, the model is optimized through a decoupled framework, reducing computational complexity during the training process and enhancing scalability. Extensive experiments demonstrate that Lighter-X achieves comparable performance to baseline models with significantly fewer parameters. In particular, on large-scale interaction graphs with millions of edges, we are able to attain even better results with only 1% of the parameter over LightGCN.\n          <\/jats:p>","DOI":"10.14778\/3749646.3749649","type":"journal-article","created":{"date-parts":[[2025,9,4]],"date-time":"2025-09-04T17:55:06Z","timestamp":1757008506000},"page":"3721-3729","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Lighter-X: An Efficient and Plug-and-Play Strategy for Graph-Based Recommendation through Decoupled Propagation"],"prefix":"10.14778","volume":"18","author":[{"given":"Yanping","family":"Zheng","sequence":"first","affiliation":[{"name":"Renmin University of China"}]},{"given":"Zhewei","family":"Wei","sequence":"additional","affiliation":[{"name":"Renmin University of China"}]},{"given":"Frank","family":"de Hoog","sequence":"additional","affiliation":[{"name":"Data 61, CSIRO"}]},{"given":"Xu","family":"Chen","sequence":"additional","affiliation":[{"name":"Renmin University of China"}]},{"given":"Hongteng","family":"Xu","sequence":"additional","affiliation":[{"name":"Renmin University of China"}]},{"given":"Yuhang","family":"Ye","sequence":"additional","affiliation":[{"name":"Huawei Poisson Lab"}]},{"given":"Jiadeng","family":"Huang","sequence":"additional","affiliation":[{"name":"Huawei Poisson Lab"}]}],"member":"320","published-online":{"date-parts":[[2025,9,4]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Divya Mahajan, and Prashant J Nair.","author":"Adnan Muhammad","year":"2021","unstructured":"Muhammad Adnan, Yassaman Ebrahimzadeh Maboud, Divya Mahajan, and Prashant J Nair. 2021. Accelerating recommendation system training by leveraging popular choices. In VLDB. 127\u2013140."},{"doi-asserted-by":"crossref","unstructured":"Aleksandar Bojchevski Johannes Gasteiger Bryan Perozzi Amol Kapoor Martin Blais Benedek R\u00f3zemberczki Michal Lukasik and Stephan G\u00fcnnemann. 2020. Scaling graph neural networks with approximate pagerank. In KDD. 2464\u20132473.","key":"e_1_2_1_2_1","DOI":"10.1145\/3394486.3403296"},{"unstructured":"Xuheng Cai Chao Huang Lianghao Xia and Xubin Ren. 2023. LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation. In ICLR. https:\/\/openreview.net\/forum?id=FKXVK9dyMM","key":"e_1_2_1_3_1"},{"doi-asserted-by":"publisher","key":"e_1_2_1_4_1","DOI":"10.1109\/ICASSP.2008.4518373"},{"key":"e_1_2_1_5_1","first-page":"1","article-title":"Efficient Tree-SVD for Subset Node Embedding over Large Dynamic Graphs","volume":"1","author":"Du Xinyu","year":"2023","unstructured":"Xinyu Du, Xingyi Zhang, Sibo Wang, and Zengfeng Huang. 2023. Efficient Tree-SVD for Subset Node Embedding over Large Dynamic Graphs. PACMMOD 1, 1 (2023), 1\u201326.","journal-title":"PACMMOD"},{"volume-title":"An invitation to compressive sensing","author":"Foucart Simon","unstructured":"Simon Foucart, Holger Rauhut, Simon Foucart, and Holger Rauhut. 2013. An invitation to compressive sensing. Springer.","key":"e_1_2_1_6_1"},{"key":"e_1_2_1_7_1","first-page":"1","article-title":"A survey of graph neural networks for recommender systems: Challenges, methods, and directions","volume":"1","author":"Gao Chen","year":"2023","unstructured":"Chen Gao, Yu Zheng, Nian Li, Yinfeng Li, Yingrong Qin, Jinghua Piao, Yuhan Quan, Jianxin Chang, Depeng Jin, Xiangnan He, et al. 2023. A survey of graph neural networks for recommender systems: Challenges, methods, and directions. TORS 1, 1 (2023), 1\u201351.","journal-title":"TORS"},{"doi-asserted-by":"crossref","unstructured":"Jiayan Guo Lun Du Xu Chen Xiaojun Ma Qiang Fu Shi Han Dongmei Zhang and Yan Zhang. 2023. On Manipulating Signals of User-Item Graph: A Jacobi Polynomial-based Graph Collaborative Filtering. In KDD. 602\u2013613.","key":"e_1_2_1_8_1","DOI":"10.1145\/3580305.3599450"},{"doi-asserted-by":"publisher","key":"e_1_2_1_9_1","DOI":"10.1145\/3397271.3401063"},{"doi-asserted-by":"crossref","unstructured":"Xiangnan He Lizi Liao Hanwang Zhang Liqiang Nie Xia Hu and Tat-Seng Chua. 2017. Neural collaborative filtering. In WWW. 173\u2013182.","key":"e_1_2_1_10_1","DOI":"10.1145\/3038912.3052569"},{"unstructured":"Thomas N Kipf and Max Welling. 2016. Semi-Supervised Classification with Graph Convolutional Networks. In ICLR.","key":"e_1_2_1_11_1"},{"doi-asserted-by":"publisher","key":"e_1_2_1_12_1","DOI":"10.1145\/3219819.3220023"},{"doi-asserted-by":"publisher","key":"e_1_2_1_13_1","DOI":"10.1016\/j.jat.2019.04.005"},{"doi-asserted-by":"publisher","key":"e_1_2_1_14_1","DOI":"10.1609\/aaai.v38i8.28737"},{"doi-asserted-by":"crossref","unstructured":"Shaowen Peng Kazunari Sugiyama and Tsunenori Mine. 2022. SVD-GCN: A simplified graph convolution paradigm for recommendation. In CIKM. 1625\u20131634.","key":"e_1_2_1_15_1","DOI":"10.1145\/3511808.3557462"},{"key":"e_1_2_1_16_1","volume-title":"PolyCF: Towards the Optimal Spectral Graph Filters for Collaborative Filtering. arXiv preprint arXiv:2401.12590","author":"Qin Yifang","year":"2024","unstructured":"Yifang Qin, Wei Ju, Xiao Luo, Yiyang Gu, and Ming Zhang. 2024. PolyCF: Towards the Optimal Spectral Graph Filters for Collaborative Filtering. arXiv preprint arXiv:2401.12590 (2024)."},{"key":"e_1_2_1_17_1","volume-title":"Movie Recommender System Based On Heterogeneous Graph Neural Networks. In 2022 8th International Conference on Systems and Informatics (ICSAI). IEEE, 1\u20137.","author":"Rahman Khalil Ur","year":"2022","unstructured":"Khalil Ur Rahman, Huifang Ma, Ali Arshad, and Azad Khan Baheer. 2022. Movie Recommender System Based On Heterogeneous Graph Neural Networks. In 2022 8th International Conference on Systems and Informatics (ICSAI). IEEE, 1\u20137."},{"key":"e_1_2_1_18_1","volume-title":"BPR: Bayesian personalized ranking from implicit feedback. arXiv preprint arXiv:1205.2618","author":"Rendle Steffen","year":"2012","unstructured":"Steffen Rendle, Christoph Freudenthaler, Zeno Gantner, and Lars Schmidt-Thieme. 2012. BPR: Bayesian personalized ranking from implicit feedback. arXiv preprint arXiv:1205.2618 (2012)."},{"doi-asserted-by":"publisher","key":"e_1_2_1_19_1","DOI":"10.1007\/s11704-024-3803-z"},{"doi-asserted-by":"publisher","key":"e_1_2_1_20_1","DOI":"10.1145\/3534678.3539192"},{"doi-asserted-by":"crossref","unstructured":"Changxin Tian Yuexiang Xie Yaliang Li Nan Yang and Wayne Xin Zhao. 2022. Learning to denoise unreliable interactions for graph collaborative filtering. In SIGIR. 122\u2013132.","key":"e_1_2_1_21_1","DOI":"10.1145\/3477495.3531889"},{"key":"e_1_2_1_22_1","volume-title":"Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1376\u20131385","author":"Tian Zhen","year":"2023","unstructured":"Zhen Tian, Ting Bai, Wayne Xin Zhao, Ji-Rong Wen, and Zhao Cao. 2023. Euler-Net: Adaptive Feature Interaction Learning via Euler's Formula for CTR Prediction. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1376\u20131385."},{"doi-asserted-by":"crossref","unstructured":"Chenyang Wang Yuanqing Yu Weizhi Ma Min Zhang Chong Chen Yiqun Liu and Shaoping Ma. 2022. Towards representation alignment and uniformity in collaborative filtering. In KDD. 1816\u20131825.","key":"e_1_2_1_23_1","DOI":"10.1145\/3534678.3539253"},{"doi-asserted-by":"crossref","unstructured":"Hanzhi Wang Mingguo He Zhewei Wei Sibo Wang Ye Yuan Xiaoyong Du and Ji-Rong Wen. 2021. Approximate graph propagation. In KDD. 1686\u20131696.","key":"e_1_2_1_24_1","DOI":"10.1145\/3447548.3467243"},{"doi-asserted-by":"crossref","unstructured":"Xiang Wang Xiangnan He Meng Wang Fuli Feng and Tat-Seng Chua. 2019. Neural graph collaborative filtering. In SIGIR. 165\u2013174.","key":"e_1_2_1_25_1","DOI":"10.1145\/3331184.3331267"},{"doi-asserted-by":"crossref","unstructured":"Xiang Wang Hongye Jin An Zhang Xiangnan He Tong Xu and Tat-Seng Chua. 2020. Disentangled graph collaborative filtering. In SIGIR. 1001\u20131010.","key":"e_1_2_1_26_1","DOI":"10.1145\/3397271.3401137"},{"doi-asserted-by":"publisher","key":"e_1_2_1_27_1","DOI":"10.1016\/j.knosys.2023.110806"},{"doi-asserted-by":"publisher","key":"e_1_2_1_28_1","DOI":"10.1109\/ICDM54844.2022.00063"},{"unstructured":"Felix Wu Amauri Souza Tianyi Zhang Christopher Fifty Tao Yu and Kilian Weinberger. 2019. Simplifying graph convolutional networks. In ICML. PMLR 6861\u20136871.","key":"e_1_2_1_29_1"},{"doi-asserted-by":"crossref","unstructured":"Jiancan Wu Xiang Wang Fuli Feng Xiangnan He Liang Chen Jianxun Lian and Xing Xie. 2021. Self-supervised graph learning for recommendation. In SIGIR. 726\u2013735.","key":"e_1_2_1_30_1","DOI":"10.1145\/3404835.3462862"},{"key":"e_1_2_1_31_1","first-page":"1","article-title":"Graph neural networks in recommender systems: a survey","volume":"55","author":"Wu Shiwen","year":"2022","unstructured":"Shiwen Wu, Fei Sun, Wentao Zhang, Xu Xie, and Bin Cui. 2022. Graph neural networks in recommender systems: a survey. CSUR 55, 5 (2022), 1\u201337.","journal-title":"CSUR"},{"key":"e_1_2_1_32_1","volume-title":"Neil Shah, and Tong Zhao.","author":"Wu Xinyi","year":"2024","unstructured":"Xinyi Wu, Donald Loveland, Runjin Chen, Yozen Liu, Xin Chen, Leonardo Neves, Ali Jadbabaie, Clark Mingxuan Ju, Neil Shah, and Tong Zhao. 2024. GraphHash: Graph Clustering Enables Parameter Efficiency in Recommender Systems. arXiv preprint arXiv:2412.17245 (2024)."},{"doi-asserted-by":"publisher","key":"e_1_2_1_33_1","DOI":"10.1109\/TIP.2014.2365720"},{"doi-asserted-by":"crossref","unstructured":"Junliang Yu Hongzhi Yin Xin Xia Tong Chen Lizhen Cui and Quoc Viet Hung Nguyen. 2022. Are graph augmentations necessary? simple graph contrastive learning for recommendation. In SIGIR. 1294\u20131303.","key":"e_1_2_1_34_1","DOI":"10.1145\/3477495.3531937"},{"doi-asserted-by":"publisher","key":"e_1_2_1_35_1","DOI":"10.1145\/3589334.3645486"},{"doi-asserted-by":"publisher","key":"e_1_2_1_36_1","DOI":"10.1145\/3627673.3679773"}],"container-title":["Proceedings of the VLDB Endowment"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.14778\/3749646.3749649","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,5]],"date-time":"2025-09-05T03:21:18Z","timestamp":1757042478000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.14778\/3749646.3749649"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":36,"journal-issue":{"issue":"11","published-print":{"date-parts":[[2025,7]]}},"alternative-id":["10.14778\/3749646.3749649"],"URL":"https:\/\/doi.org\/10.14778\/3749646.3749649","relation":{},"ISSN":["2150-8097"],"issn-type":[{"type":"print","value":"2150-8097"}],"subject":[],"published":{"date-parts":[[2025,7]]},"assertion":[{"value":"2025-09-04","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}