Aug 2026· Knowledge Discovery and Data Mining· pp. 5591-5601· 0 citations· 55 references
Computer Science
TL;DR
An information-theoretic analysis is introduced, which decomposes next-item predictive information into three components: shared information, semantic-unique information, and collaborative-unique information, which proposes RIDRec, a novel role-aware sequential recommendation framework.
Abstract
Sequential recommendation benefits from incorporating item textual semantics in addition to interaction-based collaborative signals. However, many existing methods simply concatenate or jointly encode semantic and collaborative features, without explicitly modeling how each source contributes to next-item prediction. As a result, predictive signals with fundamentally different roles are often entangled in unified representations, which limits the model's ability to exploit fine-grained predictive information. In this paper, we revisit semantics-enhanced sequential recommendation by introducing an information-theoretic analysis, which decomposes next-item predictive information into three components: shared information, semantic-unique information, and collaborative-unique information. Each component plays a distinct role as predictive evidence and contributes differently to user preference modeling. Guided by this analysis, we propose RIDRec, a novel role-aware sequential recommendation framework. RIDRec learns three role-specialized representations separately through alignment and separation constraints, each independently modeling user preferences, which preserves role distinctiveness and prevents information leakage across roles. Furthermore, RIDRec employs a role-aware fusion mechanism that adaptively combines the three role-specialized representations based on their relevance to the current user context. Extensive experiments on four public benchmarks demonstrate the effectiveness of RIDRec, showing significant improvements over various existing sequential recommendation methods. Our code is available at https://github.com/jiangnanx129/RIDRec.
Personalized recommendation depends on learning semantic and collaborative relations among items from online content and user interactions. Generative recommendation (GR) offers a promising paradigm by autoregressively generating discrete semantic IDs (SIDs) of items. However, randomly initialized SID token embeddings...
Yu-Xi Huang, Si-Lin Zhou, Chenhao Wang et al.· 0 citations
Multimodal sequential recommendation (MSR) leverages textual and visual content of items to capture the evolution of user interests over time. Prior efforts have largely concentrated on multimodal fusion strategies or downstream sequential modeling, while the question of how to obtain item representations genuinely ali...
Shu-Jie Li, Guang-Hu Yuan, Jin-Feng Xu et al.· Proceedings of the 35th ACM...· 0 citations
Multimodal graph recommendation integrates observed interactions with item images and text to learn user preferences. These inputs constitute Semantic-Behavioral Evidence (SBE). Many existing graph methods incorporate collaborative and multimodal semantic relations, yet their aggregation and state updates remain contex...
Ming-Qiao Zhang, Qi-Yao Peng, Yu-Meng Wang et al.· 0 citations
This study addresses cross-domain recommendation when source-domain users and target-domain items are structurally disjoint. We propose an adaptive framework that embeds source behavior and target-item information in a shared semantic space. Its shared core combines semantic similarity, keyword mapping, preference comp...
J. Lee, K. Park, H. Kim· Applied Sciences· 0 citations
Recommender systems leveraging generative models often generate item identifiers directly, rather than ranking catalog items by a recommendation score. Recent work extends beyond pure sequential interaction signals by incorporating item content and structured relationships among items, with two distinct directions emer...
Justin Hangöbl, Marta Moscati, Alessandro B. Melchiorre et al.· Proceedings of the 35th ACM...· 0 citations
Sequential recommender systems commonly learn item representations from interaction data. This paradigm works well for frequent items, but becomes unreliable for long-tail items whose ID embeddings receive limited updates. Offline semantic embeddings derived from large language models provide useful item-side knowledge...
Song Yu, Yan-Bo Wang, L. Cao et al.· Proceedings of the 35th ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.