UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918508853460992 |
|---|---|
| author | Zhang, Hongwei Zhong, Qiqiang Cao, Jiangxia Lv, Yiyang Wang, Huanjie Guan, Liwei Yao, Jing Wang, Yiyu Shu, Junfeng Liu, Zhaojie Li, Han |
| author_facet | Zhang, Hongwei Zhong, Qiqiang Cao, Jiangxia Lv, Yiyang Wang, Huanjie Guan, Liwei Yao, Jing Wang, Yiyu Shu, Junfeng Liu, Zhaojie Li, Han |
| contents | Modeling ultra-long user sequences involves a difficult trade-off between efficiency and effectiveness. While current paradigms rely on either item-specific search or item-agnostic compression, we propose UxSID, a framework exploring a third path: semantic-group shared interest memory. By utilizing Semantic IDs (SIDs) and a dual-level attention strategy, UxSID captures target-aware preferences without the heavy cost of item-specific models. This end-to-end architecture balances computational parsimony with semantic awareness, achieving state-of-the-art performance and a 0.337% revenue lift in large-scale advertising A/B test. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_09040 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence Zhang, Hongwei Zhong, Qiqiang Cao, Jiangxia Lv, Yiyang Wang, Huanjie Guan, Liwei Yao, Jing Wang, Yiyu Shu, Junfeng Liu, Zhaojie Li, Han Artificial Intelligence Information Retrieval Machine Learning Modeling ultra-long user sequences involves a difficult trade-off between efficiency and effectiveness. While current paradigms rely on either item-specific search or item-agnostic compression, we propose UxSID, a framework exploring a third path: semantic-group shared interest memory. By utilizing Semantic IDs (SIDs) and a dual-level attention strategy, UxSID captures target-aware preferences without the heavy cost of item-specific models. This end-to-end architecture balances computational parsimony with semantic awareness, achieving state-of-the-art performance and a 0.337% revenue lift in large-scale advertising A/B test. |
| title | UxSID: Semantic-Aware User Interests Modeling for Ultra-Long Sequence |
| topic | Artificial Intelligence Information Retrieval Machine Learning |
| url | https://arxiv.org/abs/2605.09040 |