S$^2$GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Guo, Zihao, Wang, Jian, Zhou, Ruxin, Liu, Youhua, Guo, Jiawei, Zhao, Jun, Xu, Xiaoxiao, Liu, Yongqi, Zhan, Kaiqiao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918508551471104
author Guo, Zihao
Wang, Jian
Zhou, Ruxin
Liu, Youhua
Guo, Jiawei
Zhao, Jun
Xu, Xiaoxiao
Liu, Yongqi
Zhan, Kaiqiao
author_facet Guo, Zihao
Wang, Jian
Zhou, Ruxin
Liu, Youhua
Guo, Jiawei
Zhao, Jun
Xu, Xiaoxiao
Liu, Yongqi
Zhan, Kaiqiao
contents Generative Recommendation (GR) has emerged as a transformative paradigm with its end-to-end generation advantages. However, existing GR methods primarily focus on direct Semantic ID (SID) generation from interaction sequences, failing to activate deeper reasoning capabilities analogous to those in large language models and thus limiting performance potential. We identify two critical limitations in current reasoning-enhanced GR approaches: (1) Strict sequential separation between reasoning and generation steps creates imbalanced computational focus across hierarchical SID codes, degrading quality for SID codes; (2) Generated reasoning vectors lack interpretable semantics, while reasoning paths suffer from unverifiable supervision. In this paper, we propose stepwise semantic-guided reasoning in latent space (S$^2$GR), a novel reasoning enhanced GR framework. First, we establish a robust semantic foundation via codebook optimization, integrating item co-occurrence relationship to capture behavioral patterns, and load balancing and uniformity objectives that maximize codebook utilization while reinforcing coarse-to-fine semantic hierarchies. Our core innovation introduces the stepwise reasoning mechanism inserting thinking tokens before each SID generation step, where each token explicitly represents coarse-grained semantics supervised via contrastive learning against ground-truth codebook cluster distributions ensuring physically grounded reasoning paths and balanced computational focus across all SID codes. Extensive experiments demonstrate the superiority of S$^2$GR, and online A/B test confirms efficacy on large-scale industrial short video platform.
format Preprint
id arxiv_https___arxiv_org_abs_2601_18664
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle S$^2$GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation
Guo, Zihao
Wang, Jian
Zhou, Ruxin
Liu, Youhua
Guo, Jiawei
Zhao, Jun
Xu, Xiaoxiao
Liu, Yongqi
Zhan, Kaiqiao
Information Retrieval
Generative Recommendation (GR) has emerged as a transformative paradigm with its end-to-end generation advantages. However, existing GR methods primarily focus on direct Semantic ID (SID) generation from interaction sequences, failing to activate deeper reasoning capabilities analogous to those in large language models and thus limiting performance potential. We identify two critical limitations in current reasoning-enhanced GR approaches: (1) Strict sequential separation between reasoning and generation steps creates imbalanced computational focus across hierarchical SID codes, degrading quality for SID codes; (2) Generated reasoning vectors lack interpretable semantics, while reasoning paths suffer from unverifiable supervision. In this paper, we propose stepwise semantic-guided reasoning in latent space (S$^2$GR), a novel reasoning enhanced GR framework. First, we establish a robust semantic foundation via codebook optimization, integrating item co-occurrence relationship to capture behavioral patterns, and load balancing and uniformity objectives that maximize codebook utilization while reinforcing coarse-to-fine semantic hierarchies. Our core innovation introduces the stepwise reasoning mechanism inserting thinking tokens before each SID generation step, where each token explicitly represents coarse-grained semantics supervised via contrastive learning against ground-truth codebook cluster distributions ensuring physically grounded reasoning paths and balanced computational focus across all SID codes. Extensive experiments demonstrate the superiority of S$^2$GR, and online A/B test confirms efficacy on large-scale industrial short video platform.
title S$^2$GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation
topic Information Retrieval
url https://arxiv.org/abs/2601.18664