arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.06642cs.IRcs.AIcs.CL
收藏

TRAWL:借助大语言模型的外部知识增强推荐

TRAWL: External Knowledge-Enhanced Recommendation with LLM Assistance

State Key Laboratory for Novel Software Technology, Nanjing University(南京大学计算机软件新技术国家重点实验室);Tencent(腾讯)

Weiqing Luo, Chonggang Song, Lingling Yi, Gong Cheng

AI总结:

本文提出TRAWL方法,利用大语言模型从外部数据提取知识,并通过对比学习适配器训练,以增强基于行为的推荐系统,实验验证了其有效性。

摘要:

将语义信息与行为数据相结合是推荐系统中的一个关键研究领域。一种有前景的方法是利用外部知识来为基于行为的推荐系统丰富语义信息。然而,这种方法面临两个主要挑战:对原始外部知识的去噪以及语义表示的适配。为解决这些问题,我们提出了一种借助大语言模型(LLM)的外部知识增强推荐方法(TRAWL)。该方法利用大语言模型从原始外部数据中提取相关的推荐知识,并采用对比学习策略进行适配器训练。在公开数据集和真实在线推荐系统上的实验验证了我们方法的有效性。

英文摘要:

Combining semantic information with behavioral data is a crucial research area in recommender systems. A promising approach involves leveraging external knowledge to enrich behavioral-based recommender systems with abundant semantic information. However, this approach faces two primary challenges: denoising raw external knowledge and adapting semantic representations. To address these challenges, we propose an External Knowledge-Enhanced Recommendation method with LLM Assistance (TRAWL). This method utilizes large language models (LLMs) to extract relevant recommendation knowledge from raw external data and employs a contrastive learning strategy for adapter training. Experiments on public datasets and real-world online recommender systems validate the effectiveness of our approach.

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。

↑