LISRec: Modeling User Preferences with Learned Item Shortcuts for Sequential Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xin, Haidong, Liu, Zhenghao, Mei, Sen, Yan, Yukun, Yu, Shi, Wang, Shuo, Chen, Zulong, Gu, Yu, Yu, Ge, Xiong, Chenyan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911287040016384
author Xin, Haidong
Liu, Zhenghao
Mei, Sen
Yan, Yukun
Yu, Shi
Wang, Shuo
Chen, Zulong
Gu, Yu
Yu, Ge
Xiong, Chenyan
author_facet Xin, Haidong
Liu, Zhenghao
Mei, Sen
Yan, Yukun
Yu, Shi
Wang, Shuo
Chen, Zulong
Gu, Yu
Yu, Ge
Xiong, Chenyan
contents User-item interaction histories are pivotal for sequential recommendation systems but often include noise, such as unintended clicks or actions that fail to reflect genuine user preferences. To address this, we propose Learned Item Shortcuts for Sequential Recommendation (LISRec), a novel framework that explicitly captures stable preferences by extracting personalized semantic shortcuts from historical interactions. LISRec first learns task-agnostic semantic representations to assess item similarities, then constructs a personalized semantic graph over all user-interacted items. By identifying the maximal semantic connectivity subset within this graph, LISRec selects the most representative items as semantic shortcuts to guide user preference modeling. This focused representation filters out irrelevant actions while preserving the diversity of genuine interests. Experimental results on the Yelp and Amazon Product datasets illustrate that LISRec achieves a 13% improvement over baseline recommendation models, showing its effectiveness in capturing stable user interests. Further analysis indicates that shortcut-based histories better capture user preferences, making more accurate and relevant recommendations. All codes and datasets are available at https://github.com/NEUIR/LISRec.
format Preprint
id arxiv_https___arxiv_org_abs_2505_22130
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LISRec: Modeling User Preferences with Learned Item Shortcuts for Sequential Recommendation
Xin, Haidong
Liu, Zhenghao
Mei, Sen
Yan, Yukun
Yu, Shi
Wang, Shuo
Chen, Zulong
Gu, Yu
Yu, Ge
Xiong, Chenyan
Information Retrieval
User-item interaction histories are pivotal for sequential recommendation systems but often include noise, such as unintended clicks or actions that fail to reflect genuine user preferences. To address this, we propose Learned Item Shortcuts for Sequential Recommendation (LISRec), a novel framework that explicitly captures stable preferences by extracting personalized semantic shortcuts from historical interactions. LISRec first learns task-agnostic semantic representations to assess item similarities, then constructs a personalized semantic graph over all user-interacted items. By identifying the maximal semantic connectivity subset within this graph, LISRec selects the most representative items as semantic shortcuts to guide user preference modeling. This focused representation filters out irrelevant actions while preserving the diversity of genuine interests. Experimental results on the Yelp and Amazon Product datasets illustrate that LISRec achieves a 13% improvement over baseline recommendation models, showing its effectiveness in capturing stable user interests. Further analysis indicates that shortcut-based histories better capture user preferences, making more accurate and relevant recommendations. All codes and datasets are available at https://github.com/NEUIR/LISRec.
title LISRec: Modeling User Preferences with Learned Item Shortcuts for Sequential Recommendation
topic Information Retrieval
url https://arxiv.org/abs/2505.22130