arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2402.04527cs.IRcs.AI
收藏

RA-Rec:一种用于基于大语言模型推荐的高效ID表示对齐框架

RA-Rec: An Efficient ID Representation Alignment Framework for LLM-based Recommendation

Huawei Poisson Lab(华为泊松实验室);University College London(伦敦大学学院)

Xiaohan Yu, Li Zhang, Xin Zhao, Yue Wang, Zhongrui Ma

AI总结:

针对基于LLM的推荐系统中ID使用缺乏推荐知识和唯一性的问题,提出ID表示对齐框架RA-Rec,将ID嵌入作为软提示,通过创新对齐模块和高效微调,以少10倍训练数据实现高达3.0%的HitRate@100提升。

摘要:

大型语言模型(LLM)近年来已成为处理各种自然语言处理任务的强大工具,引发了将LLM与推荐系统相结合的新浪潮,称为基于LLM的推荐系统(LLM-based RS)。当前的方法通常分为两大范式:ID直接使用范式和ID翻译范式,其核心弱点在于缺乏推荐知识和唯一性。为解决这一局限,我们提出了一种新范式——ID表示,它以互补的方式将预训练的ID嵌入融入LLM中。在本工作中,我们提出了RA-Rec,一种用于基于LLM推荐的高效ID表示对齐框架,该框架兼容多种基于ID的方法和LLM架构。具体而言,我们将ID嵌入视为软提示,设计了一个创新的对齐模块和一种高效的微调方法,并针对对齐定制了数据构建。大量实验表明,RA-Rec显著优于当前最先进的方法,在使用不到10倍训练数据的情况下,实现了高达3.0%的绝对HitRate@100提升。

英文摘要:

Large language models (LLM) have recently emerged as a powerful tool for a variety of natural language processing tasks, bringing a new surge of combining LLM with recommendation systems, termed as LLM-based RS. Current approaches generally fall into two main paradigms, the ID direct usage paradigm and the ID translation paradigm, noting their core weakness stems from lacking recommendation knowledge and uniqueness. To address this limitation, we propose a new paradigm, ID representation, which incorporates pre-trained ID embeddings into LLMs in a complementary manner. In this work, we present RA-Rec, an efficient ID representation alignment framework for LLM-based recommendation, which is compatible with multiple ID-based methods and LLM architectures. Specifically, we treat ID embeddings as soft prompts and design an innovative alignment module and an efficient tuning method with tailored data construction for alignment. Extensive experiments demonstrate RA-Rec substantially outperforms current state-of-the-art methods, achieving up to 3.0% absolute HitRate@100 improvements while utilizing less than 10x training data.

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。

↑