Xingang Wu*, Yajie Huang, Yu Li, Yifei Duan, Xiaowei Li, Mingze Wu
Yantai Institute of Technology, Yantai 265699, Shandong, China.
*Corresponding author: Xingang Wu
Abstract
This paper aims to explore and implement a question-answering system utilizing Retrieval-Augmented Generation (RAG) technology with Large Language Models (LLMs)to enhance the accuracy and reliability of information retrieval. The paper first compares and analyzes the performance characteristics and application scenarios of mainstream large models. It then systematically reviews the evolution of RAG technology and its key optimization strategies. Building on this foundation, the paper delves into the development of question-answering retrieval systems, from traditional template matching to end-to-end deep learning models, and finally to the current integration of large models and RAG. The core section of this paper designs and implements a question-answering retrieval system based on large models and RAG technology, detailing its system architecture, workflow, and key implementation technologies using LangChain and vector databases. Through the construction of experimental prototypes and comparative testing, the system’s significant advantages in answer accuracy, factual consistency, and resistance to “hallucinations” compared to pure large model solutions are verified. Finally, the paper summarizes the current system's shortcomings and looks forward to future research directions.
References
[1] Jiang P, Ouyang S, Jiao Y, Zhong M, Tian R, Han J. A Survey on Retrieval And Structuring Augmented Generation with Large Language Models. arXiv [Preprint] 2025 [cited 2025 Sep 12].
Available from: http://arxiv.org/abs/2509.10697
[2] Oche AJ, Folashade AG, Ghosal T, Biswas A. A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions. arXiv [Preprint] 2025 [cited 2025 Jul 25]. Available from: http://arxiv.org/abs/2507.18910
[3] Lewis P, Perez E, Piktus A, Petroni F, Karpukhin V, Goyal N, et al. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. arXiv [Preprint] 2020 [cited 2020 May 22]. Available from: http://arxiv.org/abs/2005.11401
[4] Chen J, Lin H, Han X, Sun L. Benchmarking Large Language Models in Retrieval-Augmented Generation. arXiv [Preprint] 2023 [cited 2023 Sep 4]. Available from:
http://arxiv.org/abs/2309.01431
[5] Danuarta L, Mawardi VC, Lee V. Retrieval-Augmented Generation (RAG) Large Language Model For Educational Chatbot. In: 2024 Ninth International Conference on Informatics and Computing (ICIC) [Internet]. Medan, Indonesia: IEEE; 2024 [cited 2025 Apr 15]. p. 1-6. Available from: https://doi.org/10.1109/ICIC64337.2024.10957676
[6] Muludi K, Fitria KM, Triloka J, Sutedi. Retrieval-Augmented Generation Approach: Document Question Answering using Large Language Model. Int J Adv Comput Sci Appl. 2024;15(3). Available from:
https://thesai.org/Publications/ViewPaper?Volume=15&Issue=3&Code=IJACSA&SerialNo=46
[7] LangChain [Internet]. [place unknown]: LangChain Inc.; c2024. LangChain overview; [cited 2024 Date]. Available from:
https://www.langchain.com/
[8] Du Z, Qian Y, Liu X, Ding M, Qiu J, Yang Z, et al. GLM: General Language Model Pretraining with Autoregressive Blank Infilling. arXiv [Preprint] 2021 [cited 2021 Mar 18]. Available from:
http://arxiv.org/abs/2103.10360
How to cite this paper
A Study on Question-answering Retrieval System Based on Large Language Models and RAG Technology
How to cite this paper: Xingang Wu, Yajie Huang, Yu Li, Yifei Duan, Xiaowei Li, Mingze Wu. (2025). A Study on Question-answering Retrieval System Based on Large Language Models and RAG Technology. Engineering Advances, 5(4), 166-171.
DOI: http://dx.doi.org/10.26855/ea.2025.10.007