{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T05:11:51Z","timestamp":1771477911166,"version":"3.50.1"},"reference-count":30,"publisher":"Oxford University Press (OUP)","issue":"2","license":[{"start":{"date-parts":[[2025,9,28]],"date-time":"2025-09-28T00:00:00Z","timestamp":1759017600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"DOI":"10.13039\/501100012325","name":"National Social Science Fund of China","doi-asserted-by":"publisher","award":["22BTQ045"],"award-info":[{"award-number":["22BTQ045"]}],"id":[{"id":"10.13039\/501100012325","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Research Fund of National Language Commission","award":["YB145-123"],"award-info":[{"award-number":["YB145-123"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,2,15]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>In the context of multi-scenario recommendation, multi-scenario click-through-rate (MS-CTR) prediction plays a crucial role in effectively personalizing recommendations on commercial platforms. However, when confronted with multi-scenario data, models often face challenges such as overfitting, inadequate feature representation, and unstable optimization, which hinder the performance and reliability of CTR prediction. To tackle these challenges, this study introduces the enhanced scene-aware transformer (ESAT) framework for MS-CTR prediction. This framework is divided into five modules. First, the structural position-aware scene encoding module converts scene attributes into fixed-dimensional embedding vectors and the scene adaptive transformation module uses a nonlinear transformation to dynamically adjust scene features. The cross-scene regularization module uses multi-sample dropout technology to enhance generalization ability and prevent overfitting. The scene-aware discriminative learning module applies contrastive learning to optimize the similarity between samples. Finally, the hierarchical stability control module introduces ClippyGrad optimizer and $L_\\infty $ regularization to accurately control gradient updates, avoid excessive steps, and improve training stability. Experiments conducted on large-scale multi-scenario datasets confirm the effectiveness of the proposed module. The results indicate that the ESAT model substantially elevates performance in MS-CTR prediction tasks, especially in terms of generalization to new scenes, precision in feature representation, and stability in parameter updates.<\/jats:p>","DOI":"10.1093\/comjnl\/bxaf112","type":"journal-article","created":{"date-parts":[[2025,9,8]],"date-time":"2025-09-08T11:23:27Z","timestamp":1757330607000},"page":"247-258","source":"Crossref","is-referenced-by-count":0,"title":["Multi-scenario CTR prediction via enhanced scene-aware transformer framework"],"prefix":"10.1093","volume":"69","author":[{"given":"Weizhong","family":"Liu","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology, Guangdong University of Technology, University Town Campus , No. 100, Outer Ring West Road, Xiaoguwei Street, Panyu District, Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qifeng","family":"Bai","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Guangdong University of Technology, University Town Campus , No. 100, Outer Ring West Road, Xiaoguwei Street, Panyu District, Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Wentao","family":"Huang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Guangdong University of Technology, University Town Campus , No. 100, Outer Ring West Road, Xiaoguwei Street, Panyu District, Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Feiyan","family":"Pang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Guangdong University of Technology, University Town Campus , No. 100, Outer Ring West Road, Xiaoguwei Street, Panyu District, Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Nankai","family":"Lin","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Guangdong University of Foreign Studies , Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Aimin","family":"Yang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Guangdong University of Technology, University Town Campus , No. 100, Outer Ring West Road, Xiaoguwei Street, Panyu District, Guangzhou 510006, Guangdong ,","place":["China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2025,9,28]]},"reference":[{"key":"2026021823462772800_ref1","author":"Guo","year":"2017"},{"key":"2026021823462772800_ref2","first-page":"2605","article-title":"Improving multi-scenario learning to rank in e-commerce by exploiting task relationships in the label space","volume-title":"Proceedings of the 29th ACM international conference on information and knowledge management, CIKM \u201820, virtual event, Ireland, 19\u201323 October","author":"Li","year":"2020"},{"key":"2026021823462772800_ref3","first-page":"4094","article-title":"SAR-Net: a scenario-aware ranking network for personalized fair recommendation in hundreds of travel scenarios","volume-title":"Proceedings of the 30th ACM international conference on information and knowledge management, CIKM \u201821, virtual event, Quebec City, Canada, 20\u201324 October","author":"Shen","year":"2021"},{"key":"2026021823462772800_ref4","first-page":"4104","article-title":"One model to serve all: star topology adaptive recommender for multi-domain CTR prediction","volume-title":"Proceedings of the 30th ACM international conference on information and knowledge management, CIKM \u201821, virtual event, Quebec City, Canada, 20\u201324 October","author":"Sheng","year":"2021"},{"key":"2026021823462772800_ref5","doi-asserted-by":"crossref","first-page":"1368","DOI":"10.1145\/3488560.3498479","article-title":"Leaving no one behind: a multi-scenario multi-task meta learning approach for advertiser modeling","volume-title":"Proceedings of the fifteenth ACM international conference on web search and data mining, WSDM \u201822, virtual event, AZ, USA, 21\u201325 February","author":"Zhang","year":"2022"},{"key":"2026021823462772800_ref6","first-page":"3674","article-title":"Scenario-adaptive and self-supervised model for multi-scenario personalized recommendation","volume-title":"Proceedings of the 31st ACM international conference on information and knowledge management, CIKM \u201822, Atlanta, GA, USA, 17\u201321 October","author":"Zhang","year":"2022"},{"key":"2026021823462772800_ref7","first-page":"1930","article-title":"Modeling task relationships in multi-task learning with multi-gate mixture-of-experts","volume-title":"Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, United Kingdom, 19\u201323 July","author":"Ma","year":"2018"},{"key":"2026021823462772800_ref8","doi-asserted-by":"crossref","first-page":"269","DOI":"10.1145\/3383313.3412236","article-title":"Progressive layered extraction (PLE): a novel multi-task learning (MTL) model for personalized recommendations","volume-title":"Proceedings of the 14th ACM conference on recommender systems, RecSys \u201820, virtual event, Brazil, 22\u201326 September","author":"Tang","year":"2020"},{"key":"2026021823462772800_ref9","article-title":"An overview of multi-task learning in deep neural networks","author":"Ruder","year":"2017"},{"key":"2026021823462772800_ref10","first-page":"1754","article-title":"xDeepFM: combining explicit and implicit feature interactions for recommender systems","volume-title":"Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2018, London, United Kingdom, 19\u201323 July","author":"Lian","year":"2018"},{"key":"2026021823462772800_ref11","first-page":"1149","article-title":"Product-based neural networks for user response prediction","volume-title":"Proceedings of IEEE 16th international conference on data mining, ICDM 2016, Barcelona, Catalonia, Spain, 12\u201315 December","author":"Qu","year":"2016"},{"key":"2026021823462772800_ref12","doi-asserted-by":"crossref","first-page":"240","DOI":"10.1145\/3383313.3412488","article-title":"Neural collaborative filtering vs. matrix factorization revisited","volume-title":"Proceedings of the 14th ACM Conference on Recommender Systems, Virtual Event, Brazil, 22\u201326 September","author":"Rendle","year":"2020"},{"key":"2026021823462772800_ref13","first-page":"355","article-title":"Neural factorization machines for sparse predictive analytics","volume-title":"Proceedings of the 40th international ACM SIGIR conference on Research and Development in information retrieval, SIGIR 2017, Tokyo, Japan, 7\u201311 August","author":"He","year":"2017"},{"key":"2026021823462772800_ref14","first-page":"1","article-title":"Deep & cross network for ad click predictions","volume-title":"Proceedings of the 1st international workshop on deep learning practice for high-dimensional sparse data, ADKDD\u201917, Halifax, NS, Canada, 14 August","author":"Wang","year":"2017"},{"key":"2026021823462772800_ref15","first-page":"127","article-title":"Scenario-aware and mutual-based approach for multi-scenario recommendation in e-commerce","volume-title":"Proceedings of the 2020 international conference on data mining workshops, ICDMW \u201820, Sorrento, Italy, 17\u201320 November","author":"Chen","year":"2020"},{"key":"2026021823462772800_ref16","first-page":"2732","article-title":"Internal and contextual attention network for cold-start multi-channel matching in recommendation","volume-title":"Proceedings of the twenty-ninth international joint conference on artificial intelligence, IJCAI \u201820, virtual event, Yokohama, Japan, 7 January\u201315 January","author":"Xie","year":"2020"},{"key":"2026021823462772800_ref17","doi-asserted-by":"crossref","first-page":"4661","DOI":"10.1145\/3580305.3599936","article-title":"Scenario-adaptive feature interaction for click-through rate prediction","volume-title":"Proceedings of the 29th ACM SIGKDD conference on knowledge discovery and data mining, KDD \u201823, Long Beach, CA, USA, 6\u201310 August","author":"Min","year":"2023"},{"key":"2026021823462772800_ref18","article-title":"Understanding scaling laws for recommendation models","author":"Ardalani","year":"2022"},{"key":"2026021823462772800_ref19","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1007\/978-1-4615-5529-2_5","article-title":"Multitask learning","volume-title":"Learning to Learn","author":"Caruana","year":"1998"},{"key":"2026021823462772800_ref20","author":"Guo"},{"key":"2026021823462772800_ref21","first-page":"805","article-title":"CL4CTR: a contrastive learning framework for CTR prediction","volume-title":"Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining, Singapore, 27 February\u20133 March","author":"Wang","year":"2023"},{"key":"2026021823462772800_ref22","first-page":"1857","article-title":"Hybrid contrastive constraints for multi-scenario ad ranking","volume-title":"Proceedings of the 32nd ACM International Conference on Information and Knowledge Management, Birmingham, United Kingdom, 21\u201325 October","author":"Mu","year":"2023"},{"key":"2026021823462772800_ref23","doi-asserted-by":"crossref","first-page":"7","DOI":"10.1145\/2988450.2988454","article-title":"Wide & deep learning for recommender systems","volume-title":"Proceedings of the 1st Workshop on Deep Learning for Recommender Systems, Boston, MA, USA, 15 September","author":"Cheng","year":"2016"},{"key":"2026021823462772800_ref24","first-page":"43","article-title":"Recommending what video to watch next: a multitask ranking system","volume-title":"Proceedings of the 13th ACM Conference on Recommender Systems, Copenhagen, Denmark, 16\u201320 September","author":"Zhao","year":"2019"},{"key":"2026021823462772800_ref25","article-title":"Multi-sample dropout for accelerated training and better generalization","author":"Inoue","year":"2019"},{"key":"2026021823462772800_ref26","first-page":"1259","article-title":"Contrastive learning for sequential recommendation","volume-title":"2022 IEEE 38th International Conference on Data Engineering (ICDE), (Virtual) Kuala Lumpur, Malaysia, 9 May","author":"Xie","year":"2022"},{"key":"2026021823462772800_ref27","first-page":"2121","article-title":"Adaptive subgradient methods for online learning and stochastic optimization","volume":"12","author":"Duchi","year":"2011","journal-title":"J Mach Learn Res"},{"key":"2026021823462772800_ref28","doi-asserted-by":"crossref","first-page":"4882","DOI":"10.1145\/3580305.3599846","article-title":"Improving training stability for multitask ranking models in recommender systems","volume-title":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Long Beach, CA, USA, 6\u201310 August","author":"Tang","year":"2023"},{"key":"2026021823462772800_ref29","author":"Xiao"},{"key":"2026021823462772800_ref30","first-page":"4635","article-title":"AdaSparse: learning adaptively sparse structures for multi-domain click-through rate prediction","volume-title":"Proceedings of the 31st ACM International Conference on Information and Knowledge Management, Atlanta, GA, USA, 17\u201321 October","author":"Yang","year":"2022"}],"container-title":["The Computer Journal"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/comjnl\/article-pdf\/69\/2\/247\/64419664\/bxaf112.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/comjnl\/article-pdf\/69\/2\/247\/64419664\/bxaf112.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T04:46:39Z","timestamp":1771476399000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/comjnl\/article\/69\/2\/247\/8267000"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,28]]},"references-count":30,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,9,28]]},"published-print":{"date-parts":[[2026,2,15]]}},"URL":"https:\/\/doi.org\/10.1093\/comjnl\/bxaf112","relation":{},"ISSN":["0010-4620","1460-2067"],"issn-type":[{"value":"0010-4620","type":"print"},{"value":"1460-2067","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,2]]},"published":{"date-parts":[[2025,9,28]]}}}