GLASS: A Generative Recommender for Long-sequence Modeling via SID-Tier and Semantic Search

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cao, Shiteng, She, Junda, Liu, Ji, Zeng, Bin, Guo, Chengcheng, Cai, Kuo, Luo, Qiang, Tang, Ruiming, Li, Han, Gai, Kun, Li, Zhiheng, Yang, Cheng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908815751905280
author Cao, Shiteng
She, Junda
Liu, Ji
Zeng, Bin
Guo, Chengcheng
Cai, Kuo
Luo, Qiang
Tang, Ruiming
Li, Han
Gai, Kun
Li, Zhiheng
Yang, Cheng
author_facet Cao, Shiteng
She, Junda
Liu, Ji
Zeng, Bin
Guo, Chengcheng
Cai, Kuo
Luo, Qiang
Tang, Ruiming
Li, Han
Gai, Kun
Li, Zhiheng
Yang, Cheng
contents Leveraging long-term user behavioral patterns is a key trajectory for enhancing the accuracy of modern recommender systems. While generative recommender systems have emerged as a transformative paradigm, they face hurdles in effectively modeling extensive historical sequences. To address this challenge, we propose GLASS, a novel framework that integrates long-term user interests into the generative process via SID-Tier and Semantic Search. We first introduce SID-Tier, a module that maps long-term interactions into a unified interest vector to enhance the prediction of the initial SID token. Unlike traditional retrieval models that struggle with massive item spaces, SID-Tier leverages the compact nature of the semantic codebook to incorporate cross features between the user's long-term history and candidate semantic codes. Furthermore, we present semantic hard search, which utilizes generated coarse-grained semantic ID as dynamic keys to extract relevant historical behaviors, which are then fused via an adaptive gated fusion module to recalibrate the trajectory of subsequent fine-grained tokens. To address the inherent data sparsity in semantic hard search, we propose two strategies: semantic neighbor augmentation and codebook resizing. Extensive experiments on two large-scale real-world datasets, TAOBAO-MM and KuaiRec, demonstrate that GLASS outperforms state-of-the-art baselines, achieving significant gains in recommendation quality. Our codes are made publicly available to facilitate further research in generative recommendation.
format Preprint
id arxiv_https___arxiv_org_abs_2602_05663
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle GLASS: A Generative Recommender for Long-sequence Modeling via SID-Tier and Semantic Search
Cao, Shiteng
She, Junda
Liu, Ji
Zeng, Bin
Guo, Chengcheng
Cai, Kuo
Luo, Qiang
Tang, Ruiming
Li, Han
Gai, Kun
Li, Zhiheng
Yang, Cheng
Information Retrieval
Leveraging long-term user behavioral patterns is a key trajectory for enhancing the accuracy of modern recommender systems. While generative recommender systems have emerged as a transformative paradigm, they face hurdles in effectively modeling extensive historical sequences. To address this challenge, we propose GLASS, a novel framework that integrates long-term user interests into the generative process via SID-Tier and Semantic Search. We first introduce SID-Tier, a module that maps long-term interactions into a unified interest vector to enhance the prediction of the initial SID token. Unlike traditional retrieval models that struggle with massive item spaces, SID-Tier leverages the compact nature of the semantic codebook to incorporate cross features between the user's long-term history and candidate semantic codes. Furthermore, we present semantic hard search, which utilizes generated coarse-grained semantic ID as dynamic keys to extract relevant historical behaviors, which are then fused via an adaptive gated fusion module to recalibrate the trajectory of subsequent fine-grained tokens. To address the inherent data sparsity in semantic hard search, we propose two strategies: semantic neighbor augmentation and codebook resizing. Extensive experiments on two large-scale real-world datasets, TAOBAO-MM and KuaiRec, demonstrate that GLASS outperforms state-of-the-art baselines, achieving significant gains in recommendation quality. Our codes are made publicly available to facilitate further research in generative recommendation.
title GLASS: A Generative Recommender for Long-sequence Modeling via SID-Tier and Semantic Search
topic Information Retrieval
url https://arxiv.org/abs/2602.05663