{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,12]],"date-time":"2025-12-12T13:10:08Z","timestamp":1765545008378,"version":"build-2065373602"},"publisher-location":"New York, NY, USA","reference-count":43,"publisher":"ACM","license":[{"start":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T00:00:00Z","timestamp":1746662400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2025,5,8]]},"DOI":"10.1145\/3701716.3715223","type":"proceedings-article","created":{"date-parts":[[2025,5,23]],"date-time":"2025-05-23T16:20:01Z","timestamp":1748017201000},"page":"344-353","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation"],"prefix":"10.1145","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6779-2418","authenticated-orcid":false,"given":"Mingfu","family":"Liang","sequence":"first","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2336-8417","authenticated-orcid":false,"given":"Xi","family":"Liu","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8797-4646","authenticated-orcid":false,"given":"Rong","family":"Jin","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-2540-0712","authenticated-orcid":false,"given":"Boyang","family":"Liu","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8072-6060","authenticated-orcid":false,"given":"Qiuling","family":"Suo","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2571-5796","authenticated-orcid":false,"given":"Qinghai","family":"Zhou","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-6850-0614","authenticated-orcid":false,"given":"Song","family":"Zhou","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-6384-3810","authenticated-orcid":false,"given":"Laming","family":"Chen","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9555-7132","authenticated-orcid":false,"given":"Hua","family":"Zheng","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-4325-3883","authenticated-orcid":false,"given":"Zhiyuan","family":"Li","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-7999-6029","authenticated-orcid":false,"given":"Shali","family":"Jiang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-5946-5456","authenticated-orcid":false,"given":"Jiyan","family":"Yang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-5069-6298","authenticated-orcid":false,"given":"Xiaozhen","family":"Xia","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9714-2465","authenticated-orcid":false,"given":"Fan","family":"Yang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8598-2554","authenticated-orcid":false,"given":"Yasmine","family":"Badr","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8229-2294","authenticated-orcid":false,"given":"Ellie","family":"Wen","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-0901-8315","authenticated-orcid":false,"given":"Shuyu","family":"Xu","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8173-3737","authenticated-orcid":false,"given":"Hansey","family":"Chen","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-9287-9756","authenticated-orcid":false,"given":"Zhengyu","family":"Zhang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3355-8962","authenticated-orcid":false,"given":"Jade","family":"Nie","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-1681-285X","authenticated-orcid":false,"given":"Chunzhi","family":"Yang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5534-3401","authenticated-orcid":false,"given":"Zhichen","family":"Zeng","sequence":"additional","affiliation":[{"name":"University of Illinois Urbana-Champaign, Champaign, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-9380-1010","authenticated-orcid":false,"given":"Weilin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-8629-0864","authenticated-orcid":false,"given":"Xingliang","family":"Huang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2566-4316","authenticated-orcid":false,"given":"Qianru","family":"Li","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6009-1405","authenticated-orcid":false,"given":"Shiquan","family":"Wang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1349-5201","authenticated-orcid":false,"given":"Evelyn","family":"Lyu","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4730-4104","authenticated-orcid":false,"given":"Wenjing","family":"Lu","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4071-8829","authenticated-orcid":false,"given":"Rui","family":"Zhang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-6451-809X","authenticated-orcid":false,"given":"Wenjun","family":"Wang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-5237-6179","authenticated-orcid":false,"given":"Jason","family":"Rudy","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4125-2135","authenticated-orcid":false,"given":"Mengyue","family":"Hang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-5102-6217","authenticated-orcid":false,"given":"Kai","family":"Wang","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2489-200X","authenticated-orcid":false,"given":"Bo","family":"Long","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-2843-9290","authenticated-orcid":false,"given":"Wenlin","family":"Chen","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-0054-9898","authenticated-orcid":false,"given":"Santanu","family":"Kolay","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7855-3522","authenticated-orcid":false,"given":"Huayu","family":"Li","sequence":"additional","affiliation":[{"name":"Meta Platforms, Menlo Park, USA"}]}],"member":"320","published-online":{"date-parts":[[2025,5,23]]},"reference":[{"key":"e_1_3_2_2_1_1","unstructured":"[n. d.]. Kuaishou howpublished = https:\/\/www.kuaishou.com\/activity\/uimc."},{"key":"e_1_3_2_2_2_1","volume-title":"Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al.","author":"Achiam Josh","year":"2023","unstructured":"Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023)."},{"key":"e_1_3_2_2_3_1","unstructured":"Rohan Anil Sandra Gadanho Da Huang Nijith Jacob Zhuoshu Li Dong Lin Todd Phillips Cristina Pop Kevin Regan Gil I Shamir et al. 2022. On the factory floor: ML engineering for industrial-scale ads recommendation models. arXiv preprint arXiv:2209.05310 (2022)"},{"key":"e_1_3_2_2_4_1","unstructured":"Apache. 2023. ZooKeeper. https:\/\/zookeeper.apache.org\/"},{"key":"e_1_3_2_2_5_1","unstructured":"Peter Bartlett Dave Helmbold and Philip Long. 2018. Gradient descent with identity initialization efficiently learns positive definite linear transformations by deep residual networks. In ICML. PMLR 521--530."},{"key":"e_1_3_2_2_6_1","doi-asserted-by":"crossref","unstructured":"Gang Chen Jiawei Chen Fuli Feng Sheng Zhou and Xiangnan He. 2023. Unbiased knowledge distillation for recommendation. In WSDM. 976--984.","DOI":"10.1145\/3539597.3570477"},{"key":"e_1_3_2_2_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3626772.3657743"},{"key":"e_1_3_2_2_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/3523227.3546767"},{"key":"e_1_3_2_2_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-021-01453-z"},{"key":"e_1_3_2_2_10_1","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2017\/239"},{"key":"e_1_3_2_2_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/2872427.2883037"},{"key":"e_1_3_2_2_12_1","doi-asserted-by":"publisher","DOI":"10.1145\/2648584.2648589"},{"key":"e_1_3_2_2_13_1","volume-title":"Distilling the Knowledge in a Neural Network. arXiv preprint arXiv:1503.02531","author":"Hinton Geoffrey","year":"2015","unstructured":"Geoffrey Hinton. 2015. Distilling the Knowledge in a Neural Network. arXiv preprint arXiv:1503.02531 (2015)."},{"key":"e_1_3_2_2_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2021.08.060"},{"key":"e_1_3_2_2_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/3543507.3583209"},{"key":"e_1_3_2_2_16_1","volume-title":"ACM Transactions on Recommender Systems","author":"Kang SeongKu","year":"2024","unstructured":"SeongKu Kang, Wonbin Kweon, Dongha Lee, Jianxun Lian, Xing Xie, and Hwanjo Yu. 2024. Unbiased, Effective, and Efficient Distillation from Heterogeneous Models for Recommender Systems. ACM Transactions on Recommender Systems (2024)."},{"key":"e_1_3_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3640457.3688055"},{"key":"e_1_3_2_2_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3442381.3449878"},{"key":"e_1_3_2_2_19_1","doi-asserted-by":"crossref","unstructured":"Gyuseok Lee SeongKu Kang Wonbin Kweon and Hwanjo Yu. 2024. Continual Collaborative Distillation for Recommender System. In SIGKDD. 1495--1505.","DOI":"10.1145\/3637528.3671924"},{"key":"e_1_3_2_2_20_1","unstructured":"Zeyu Li Wei Cheng Yang Chen Haifeng Chen and Wei Wang. [n. d.]. Interpretable click-through rate prediction through hierarchical attention."},{"key":"e_1_3_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3523227.3551475"},{"key":"e_1_3_2_2_22_1","volume-title":"Jianyu Huang, Narayanan Sundaraman, Jongsoo Park, Xiaodong Wang, Udit Gupta, CaroleJean Wu, Alisson G Azzolini, et al.","author":"Naumov Maxim","year":"2019","unstructured":"Maxim Naumov, Dheevatsa Mudigere, Hao-Jun Michael Shi, Jianyu Huang, Narayanan Sundaraman, Jongsoo Park, Xiaodong Wang, Udit Gupta, CaroleJean Wu, Alisson G Azzolini, et al. 2019. Deep learning recommendation model for personalization and recommendation systems. arXiv preprint arXiv:1906.00091 (2019)."},{"key":"e_1_3_2_2_23_1","unstructured":"Aviv Navon. 2024. Parameter Sharing in Deep Learning. https:\/\/avivnavon.github. io\/blog\/parameter-sharing-in-deep-learning\/ Accessed: 2024--11--18."},{"key":"e_1_3_2_2_24_1","doi-asserted-by":"crossref","unstructured":"Junwei Pan Wei Xue Ximei Wang Haibin Yu Xun Liu Shijie Quan Xueming Qiu Dapeng Liu Lei Xiao and Jie Jiang. 2024. Ads recommendation in a collapsed and entangled world. In SIGKDD. 5566--5577.","DOI":"10.1145\/3637528.3671607"},{"key":"e_1_3_2_2_25_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDM.2010.127"},{"key":"e_1_3_2_2_26_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i4.25582"},{"key":"e_1_3_2_2_27_1","doi-asserted-by":"publisher","DOI":"10.1145\/3442381.3449930"},{"key":"e_1_3_2_2_28_1","doi-asserted-by":"crossref","unstructured":"Jiaxi Tang and Ke Wang. 2018. Ranking distillation: Learning compact ranking models with high performance for recommender system. In SIGKDD. 2289--2298.","DOI":"10.1145\/3219819.3220021"},{"volume-title":"d.]. Ad display\/click data on taobao.com","year":"2018","key":"e_1_3_2_2_29_1","unstructured":"Tianchi. [n. d.]. Ad display\/click data on taobao.com, 2018. https:\/\/tianchi.aliyun.com\/dataset\/56."},{"key":"e_1_3_2_2_30_1","volume-title":"Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971","author":"Touvron Hugo","year":"2023","unstructured":"Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timoth\u00e9e Lacroix, Baptiste Rozi\u00e8re, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)."},{"key":"e_1_3_2_2_31_1","doi-asserted-by":"publisher","DOI":"10.1145\/3124749.3124754"},{"key":"e_1_3_2_2_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/3442381.3450078"},{"key":"e_1_3_2_2_33_1","volume-title":"Proceedings of the 1st Adaptive & Multitask Learning Workshop.","author":"Wang Ruoxi","year":"2019","unstructured":"Ruoxi Wang, Zhe Zhao, Xinyang Yi, Ji Yang, Derek Zhiyuan Cheng, Lichan Hong, Steve Tjoa, Jieqi Kang, Evan Ettinger, and H Chi. 2019. Improving Relevance Prediction with Transfer Learning in Large-scale Retrieval Systems. In Proceedings of the 1st Adaptive & Multitask Learning Workshop."},{"key":"e_1_3_2_2_34_1","doi-asserted-by":"crossref","unstructured":"Zhibo Xiao Luwei Yang Wen Jiang Yi Wei Yi Hu and Hao Wang. 2020. Deep multi-interest network for click-through rate prediction. In CIKM. 2265--2268.","DOI":"10.1145\/3340531.3412092"},{"key":"e_1_3_2_2_35_1","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3219890"},{"key":"e_1_3_2_2_36_1","unstructured":"Zhichen Zeng Xiaolong Liu et al. 2024. InterFormer: Towards Effective Heterogeneous Interaction Learning for Click-Through Rate Prediction. arXiv preprint arXiv:2411.09852 (2024)."},{"key":"e_1_3_2_2_37_1","volume-title":"Forty-first International Conference on Machine Learning. https:\/\/openreview.net\/forum?id=8iUgr2nuwo","author":"Zhang Buyun","year":"2024","unstructured":"Buyun Zhang, Liang Luo, Yuxin Chen, Jade Nie, Xi Liu, Shen Li, Yanli Zhao, Yuchen Hao, Yantao Yao, Ellie Dingqiao Wen, Jongsoo Park, Maxim Naumov, and Wenlin Chen. 2024. Wukong: Towards a Scaling Law for Large-Scale Recommendation. In Forty-first International Conference on Machine Learning. https:\/\/openreview.net\/forum?id=8iUgr2nuwo"},{"key":"e_1_3_2_2_38_1","volume-title":"DHEN: A deep and hierarchical ensemble network for large-scale click-through rate prediction. arXiv preprint arXiv:2203.11014","author":"Zhang Buyun","year":"2022","unstructured":"Buyun Zhang, Liang Luo, Xi Liu, Jay Li, Zeliang Chen, Weilin Zhang, Xiaohan Wei, Yuchen Hao, Michael Tsang, Wenjun Wang, et al. 2022. DHEN: A deep and hierarchical ensemble network for large-scale click-through rate prediction. arXiv preprint arXiv:2203.11014 (2022)."},{"key":"e_1_3_2_2_39_1","volume-title":"Scaling User Modeling: Large-scale Online User Representations for Ads Personalization in Meta. In Companion Proceedings of the ACM on Web Conference","author":"Zhang Wei","year":"2024","unstructured":"Wei Zhang, Dai Li, Chen Liang, Fang Zhou, Zhongke Zhang, Xuewei Wang, Ru Li, Yi Zhou, Yaning Huang, Dong Liang, et al. 2024. Scaling User Modeling: Large-scale Online User Representations for Ads Personalization in Meta. In Companion Proceedings of the ACM on Web Conference 2024. 47--55."},{"key":"e_1_3_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3477495.3531723"},{"key":"e_1_3_2_2_41_1","doi-asserted-by":"crossref","unstructured":"Jieming Zhu Jinyang Liu Weiqi Li Jincai Lai Xiuqiang He Liang Chen and Zibin Zheng. 2020. Ensembled CTR prediction via knowledge distillation. In CIKM. 2941--2958.","DOI":"10.1145\/3340531.3412704"},{"key":"e_1_3_2_2_42_1","doi-asserted-by":"publisher","unstructured":"Jieming Zhu Jinyang Liu Shuai Yang Qi Zhang and Xiuqiang He. 2021. Open Benchmarking for Click-Through Rate Prediction Gianluca Demartini Guido Zuccon J. Shane Culpepper Zi Huang and Hanghang Tong (Eds.). ACM 2759-- 2769. doi:10.1145\/3459637.3482486","DOI":"10.1145\/3459637.3482486"},{"key":"e_1_3_2_2_43_1","doi-asserted-by":"publisher","DOI":"10.1145\/3589334.3645347"}],"event":{"name":"WWW '25: The ACM Web Conference 2025","sponsor":["SIGWEB ACM Special Interest Group on Hypertext, Hypermedia, and Web"],"location":"Sydney NSW Australia","acronym":"WWW '25"},"container-title":["Companion Proceedings of the ACM on Web Conference 2025"],"original-title":[],"link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3701716.3715223","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3701716.3715223","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,8]],"date-time":"2025-10-08T03:03:40Z","timestamp":1759892620000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3701716.3715223"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,8]]},"references-count":43,"alternative-id":["10.1145\/3701716.3715223","10.1145\/3701716"],"URL":"https:\/\/doi.org\/10.1145\/3701716.3715223","relation":{},"subject":[],"published":{"date-parts":[[2025,5,8]]},"assertion":[{"value":"2025-05-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}