An Integrative Approach to LLM Literature with the Combination of QLoRa, SFT and Agentic RAG

Authors : Aslı Güngör, Büşra Nur Emir, Sedanur Yılmaz, Melike Akdağ, Ali Berkol

Pages : 249-261

View : 310 | Download : 700

Publication Date : 2025-11-30

Article Type : Research Paper

Abstract :This study offers a solution for document-based question-answering systems for both mobile and web-based applications. This solution combines the fine-tuning of the transformer architecture found suitable for the problem using the appropriate dataset and the agentic Retrieval-Augmented Generation (RAG) methodology. This allows the system to handle not only document-based questions but also non-document questions through the web search agent. A separate agent structure was also incorporated into the solution to facilitate communication with the model in various languages. In the first phase, the Llama 3.1–8B Instruct model was quantized using the quantized Low-Rank Adaptation (QLoRa) method using a dataset with a context-question-answer structure and trained with Supervised Fine-Tuning (SFT). To overcome common problems encountered in the classical RAG architecture, such as hallucination existence, inaccurate document analysis, and missing answers due to insufficient context, agents such as web search, language translation, and techniques like document ranking, and hallucination checking were included, and the agentic RAG architecture was proposed. This system provides a dynamic structure, where user questions and answers are systematically orchestrated. The model\\\'s performance has been tested using metrics such as Exact Match, ROUGE-L, BLEU, and F1, and performance improvements have been observed. The test results demonstrate that the modular system achieved through agent integration significantly improves contextual accuracy.
Keywords : doğal dil işleme, büyük dil modelleri, ince ayar, denetimli ince ayar, nicelenmiş düşük dereceli uyum, ajan tabanlı geri almayla artırılmış üretim, çok adımlı akıl yürütme

ORIGINAL ARTICLE URL