arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2502.03041cs.IRcs.LG
收藏

大语言模型作为工业级推荐系统中的通用检索器

Large Language Model as Universal Retriever in Industrial-Scale Recommender System

Taobao & Tmall Group of Alibaba, China(淘宝与天猫集团(阿里巴巴))

Junguang Jiang, Yanwen Huang, Bin Liu, Xiaoyu Kong, Xinhang Li, Ziru Xu, Han Zhu, Jian Xu, Bo Zheng

AI总结:

本研究提出基于大语言模型的通用检索模型URM,通过多查询表示、矩阵分解和概率采样技术,实现工业级推荐系统中多目标的高效生成式检索,性能优于专用模型并提升广告核心指标3%。

摘要:

在现实世界的推荐系统中,不同的检索目标通常通过任务特定数据集与精心设计的模型架构来实现。我们证明Large Language Models(LLMs,大语言模型)可作为通用检索器,在生成式检索框架内处理多类目标。为在生成式检索中建模复杂的用户-物品关系,我们提出多查询表示方法。针对工业推荐系统中候选集规模极大的挑战,我们引入矩阵分解以提升模型的可学习性、判别性与可迁移性,并结合概率采样降低计算成本。最终,我们的Universal Retrieval Model(URM,通用检索模型)可根据任意给定目标,从数千万候选集中自适应生成结果集,且延迟控制在数十毫秒内。在工业级数据上的应用显示,URM在离线实验中优于为不同检索目标精心设计的专用专家模型,还使在线广告平台的核心指标提升了3%。

英文摘要:

In real-world recommender systems, different retrieval objectives are typically addressed using task-specific datasets with carefully designed model architectures. We demonstrate that Large Language Models (LLMs) can function as universal retrievers, capable of handling multiple objectives within a generative retrieval framework. To model complex user-item relationships within generative retrieval, we propose multi-query representation. To address the challenge of extremely large candidate sets in industrial recommender systems, we introduce matrix decomposition to boost model learnability, discriminability, and transferability, and we incorporate probabilistic sampling to reduce computation costs. Finally, our Universal Retrieval Model (URM) can adaptively generate a set from tens of millions of candidates based on arbitrary given objective while keeping the latency within tens of milliseconds. Applied to industrial-scale data, URM outperforms expert models elaborately designed for different retrieval objectives on offline experiments and significantly improves the core metric of online advertising platform by $3\%$.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑