Towards Personalized Federated Multi-Scenario Multi-Task Recommendation
In modern recommender systems, especially in e-commerce, predicting multiple targets such as click-through rate (CTR) and post-view conversion rate (CTCVR) is common. Multi-task recommender systems are increasingly popular in both research and practice, as they leverage shared knowledge across diverse business scenarios to enhance performance. However, emerging real-world scenarios and data privacy concerns complicate the development of a unified multi-task recommendation model.
Authors
- Yuxiang Lu (ORCID: https://orcid.org/0009-0002-6344-3880)
- Xiaofeng Gao (ORCID: https://orcid.org/0000-0003-1776-8799)
- Suizhi Huang (ORCID: https://orcid.org/0000-0003-0172-6711)
- Tsuyoshi Murata (ORCID: https://orcid.org/0000-0002-3818-7830)
- Xin Xin (ORCID: https://orcid.org/0000-0001-6116-9115)
- Yue Ding (ORCID: https://orcid.org/0000-0002-2911-1244)
- Xun Cai (ORCID: https://orcid.org/0009-0000-9180-5717)
- Yanbiao Ji (ORCID: https://orcid.org/0009-0006-8800-9561)
- Hongtao Lu (ORCID: https://orcid.org/0000-0003-2300-3039)
- Chang Liu (ORCID: https://orcid.org/0000-0001-9341-6002)
Institutions
- Shandong University (CN)
- Shanghai Jiao Tong University (CN)
- Institute of Science Tokyo (JP)
Publication Details
- Published
- 2025-02-26
- DOI
- https://doi.org/10.1145/3701551.3703523
- Citations
- 1
- Primary Topic
- Recommender Systems and Techniques
- Type
- article
- Field-Weighted Citation Impact
- 3.25