Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jia, Jian, Wang, Yipei, Li, Yan, Chen, Honggang, Bai, Xuehan, Liu, Zhaocheng, Liang, Jian, Chen, Quan, Li, Han, Jiang, Peng, Gai, Kun
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2405.03988
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910763187175424
author Jia, Jian
Wang, Yipei
Li, Yan
Chen, Honggang
Bai, Xuehan
Liu, Zhaocheng
Liang, Jian
Chen, Quan
Li, Han
Jiang, Peng
Gai, Kun
author_facet Jia, Jian
Wang, Yipei
Li, Yan
Chen, Honggang
Bai, Xuehan
Liu, Zhaocheng
Liang, Jian
Chen, Quan
Li, Han
Jiang, Peng
Gai, Kun
contents Contemporary recommendation systems predominantly rely on ID embedding to capture latent associations among users and items. However, this approach overlooks the wealth of semantic information embedded within textual descriptions of items, leading to suboptimal performance and poor generalizations. Leveraging the capability of large language models to comprehend and reason about textual content presents a promising avenue for advancing recommendation systems. To achieve this, we propose an Llm-driven knowlEdge Adaptive RecommeNdation (LEARN) framework that synergizes open-world knowledge with collaborative knowledge. We address computational complexity concerns by utilizing pretrained LLMs as item encoders and freezing LLM parameters to avoid catastrophic forgetting and preserve open-world knowledge. To bridge the gap between the open-world and collaborative domains, we design a twin-tower structure supervised by the recommendation task and tailored for practical industrial application. Through experiments on the real large-scale industrial dataset and online A/B tests, we demonstrate the efficacy of our approach in industry application. We also achieve state-of-the-art performance on six Amazon Review datasets to verify the superiority of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2405_03988
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LEARN: Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application
Jia, Jian
Wang, Yipei
Li, Yan
Chen, Honggang
Bai, Xuehan
Liu, Zhaocheng
Liang, Jian
Chen, Quan
Li, Han
Jiang, Peng
Gai, Kun
Information Retrieval
Artificial Intelligence
Contemporary recommendation systems predominantly rely on ID embedding to capture latent associations among users and items. However, this approach overlooks the wealth of semantic information embedded within textual descriptions of items, leading to suboptimal performance and poor generalizations. Leveraging the capability of large language models to comprehend and reason about textual content presents a promising avenue for advancing recommendation systems. To achieve this, we propose an Llm-driven knowlEdge Adaptive RecommeNdation (LEARN) framework that synergizes open-world knowledge with collaborative knowledge. We address computational complexity concerns by utilizing pretrained LLMs as item encoders and freezing LLM parameters to avoid catastrophic forgetting and preserve open-world knowledge. To bridge the gap between the open-world and collaborative domains, we design a twin-tower structure supervised by the recommendation task and tailored for practical industrial application. Through experiments on the real large-scale industrial dataset and online A/B tests, we demonstrate the efficacy of our approach in industry application. We also achieve state-of-the-art performance on six Amazon Review datasets to verify the superiority of our method.
title LEARN: Knowledge Adaptation from Large Language Model to Recommendation for Practical Industrial Application
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2405.03988