Balancing Fine-tuning and RAG: A Hybrid Strategy for Dynamic LLM Recommendation Updates
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915571565592576 |
|---|---|
| author | Meng, Changping Ling, Hongyi Wang, Jianling Liu, Yifan Zhang, Shuzhou Hong, Dapeng Gao, Mingyan Dalal, Onkar Chi, Ed Hong, Lichan Lu, Haokai Han, Ningren |
| author_facet | Meng, Changping Ling, Hongyi Wang, Jianling Liu, Yifan Zhang, Shuzhou Hong, Dapeng Gao, Mingyan Dalal, Onkar Chi, Ed Hong, Lichan Lu, Haokai Han, Ningren |
| contents | Large Language Models (LLMs) empower recommendation systems through their advanced reasoning and planning capabilities. However, the dynamic nature of user interests and content poses a significant challenge: While initial fine-tuning aligns LLMs with domain knowledge and user preferences, it fails to capture such real-time changes, necessitating robust update mechanisms. This paper investigates strategies for updating LLM-powered recommenders, focusing on the trade-offs between ongoing fine-tuning and Retrieval-Augmented Generation (RAG). Using an LLM-powered user interest exploration system as a case study, we perform a comparative analysis of these methods across dimensions like cost, agility, and knowledge incorporation. We propose a hybrid update strategy that leverages the long-term knowledge adaptation of periodic fine-tuning with the agility of low-cost RAG. We demonstrate through live A/B experiments on a billion-user platform that this hybrid approach yields statistically significant improvements in user satisfaction, offering a practical and cost-effective framework for maintaining high-quality LLM-powered recommender systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20260 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Balancing Fine-tuning and RAG: A Hybrid Strategy for Dynamic LLM Recommendation Updates Meng, Changping Ling, Hongyi Wang, Jianling Liu, Yifan Zhang, Shuzhou Hong, Dapeng Gao, Mingyan Dalal, Onkar Chi, Ed Hong, Lichan Lu, Haokai Han, Ningren Information Retrieval Large Language Models (LLMs) empower recommendation systems through their advanced reasoning and planning capabilities. However, the dynamic nature of user interests and content poses a significant challenge: While initial fine-tuning aligns LLMs with domain knowledge and user preferences, it fails to capture such real-time changes, necessitating robust update mechanisms. This paper investigates strategies for updating LLM-powered recommenders, focusing on the trade-offs between ongoing fine-tuning and Retrieval-Augmented Generation (RAG). Using an LLM-powered user interest exploration system as a case study, we perform a comparative analysis of these methods across dimensions like cost, agility, and knowledge incorporation. We propose a hybrid update strategy that leverages the long-term knowledge adaptation of periodic fine-tuning with the agility of low-cost RAG. We demonstrate through live A/B experiments on a billion-user platform that this hybrid approach yields statistically significant improvements in user satisfaction, offering a practical and cost-effective framework for maintaining high-quality LLM-powered recommender systems. |
| title | Balancing Fine-tuning and RAG: A Hybrid Strategy for Dynamic LLM Recommendation Updates |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2510.20260 |