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

TL;DR

GLASS is a novel generative recommender system that effectively models long-term user interests using SID-Tier and Semantic Search, leading to improved recommendation accuracy on large-scale datasets.

Contribution

The paper introduces SID-Tier and semantic hard search, innovative modules that enhance long-sequence modeling in generative recommenders, addressing data sparsity and scalability challenges.

Findings

GLASS outperforms state-of-the-art baselines on TAOBAO-MM and KuaiRec datasets.

Semantic hard search improves relevance by dynamically extracting pertinent historical behaviors.

Proposed strategies mitigate data sparsity issues in semantic codebook utilization.

Abstract

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.