The Efficiency vs. Accuracy Trade-off: Optimizing RAG-Enhanced LLM Recommender Systems Using Multi-Head Early Exit
Huixue Zhou, Hengrui Gu, Zaifu Zhan, Xi Liu, Kaixiong Zhou, Yongkang Xiao, Mingfu Liang, Srinivas Prasad Govindan
Abstract
The deployment of Large Language Models (LLMs) in recommender systems for Click-Through Rate (CTR) prediction requires a careful balance between computational efficiency and predictive accuracy. This paper introduces OptiRAG-Rec, a comprehensive framework that integrates Retrieval-Augmented Generation (RAG) with a novel multi-head early exit architecture to address both challenges. By leveraging Graph Convolutional Networks (GCNs) as efficient retrieval mechanisms, the framework significantly reduces data retrieval times while maintaining high model performance. Additionally, the multi-head early exit strategy dynamically terminates inference based on real-time predictive confidence assessments, enhancing responsiveness without sacrificing accuracy. Experimental results demonstrate that OptiRAG-Rec reduces computation time while preserving the precision required for reliable recommendations, establishing a new benchmark for efficient and accurate LLM deployment in recommendation.
BibTeX
@inproceedings{zhou-etal-2025-efficiency,
title = "The Efficiency vs. Accuracy Trade-off: Optimizing {RAG}-Enhanced {LLM} Recommender Systems Using Multi-Head Early Exit",
author = "Zhou, Huixue and
Gu, Hengrui and
Zhan, Zaifu and
Liu, Xi and
Zhou, Kaixiong and
Xiao, Yongkang and
Liang, Mingfu and
Govindan, Srinivas Prasad and
Chawla, Piyush and
Yang, Jiyan and
Meng, Xiangfei and
Li, Huayu and
Zhang, Buyun and
Luo, Liang and
Chen, Wen-Yen and
Han, Yiping and
Long, Bo and
Zhang, Rui and
Chen, Tianlong",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.1283/",
doi = "10.18653/v1/2025.acl-long.1283",
pages = "26443--26458",
ISBN = "979-8-89176-251-0"
}