[论文] Structuring and Tokenizing Distributed User Interest Context for Gener...

论文概要

研究领域: ML
作者: Ruizhong Qiu, Yinglong Xia, Dongqi Fu
发布时间: 2025-06-20
arXiv: 2506.16803

中文摘要

生成式推荐者，一新兴范式，已于工业推荐系统显其潜力，旨在自用户历史行为预测其下一次交互。生成式推荐之枢要在于物品令牌化，桥梁物品语义与推荐模型。然现有方法往往难同时有效组织并注入复杂用户行为与物品语义语境于推荐模型之中。

现有图基整合法（如图序列化与图神经网络）或困于规模扩展，或仅攫局部图信息。现有语义令牌化法多赖启发式，乏显式监督信号，易致不准或次优之语义表征。

为解用户兴趣语境建模之弊，本文提出G2Rec，一可扩展框架，统一整体图基用户共参与建模与语义令牌化，用于工业规模之生成式推荐。G2Rec使推荐模型能捕捉整体且具语义根基之用户兴趣原型，无需真实用户兴趣，从而于工业序列推荐中提供更全面准确之用户行为语境建模。

线上部署于诸产品界面及公共数据集上之广泛实验，证实G2Rec优于现有诸法。

原文摘要

Generative recommendation is an emerging paradigm that has shown promise in industrial recommendation systems, aiming to predict users' next interactions from their historical behaviors. At the core of generative recommendation lies item tokenization, which bridges item semantics and recommendation models. However, existing methods often struggle to effectively organize and inject complex user-behavioral and item-semantic contexts into recommendation models simultaneously. On the one hand, existing graph-based integration methods, such as graph serialization and graph neural networks, either suffer from scalability issues or exploit only local graph information. On the other hand, existing semantic tokenization methods typically rely on heuristics and lack explicit supervision signals, whi...

自动采集于 2026-06-20

#论文 #arXiv #ML #小凯