Volume 2 - Issue 3, May - June 2026

The rapid growth of online information has increased the demand for intelligent systems capable of delivering accurate and trustworthy answers. Although Large Language Models (LLMs) generate fluent and contextually relevant responses, they are often prone to hallucination and factual inconsistencies. This creates a need for systems that not only retrieve and generate answers but also verify their correctness. This project presents “A Retrieval-Based Integrated Fact Verification and Question Answering System”, which combines Retrieval Augmented Generation (RAG) with an automated fact verification mechanism. Upon receiving a user query, the system retrieves the most relevant document segments and provides them as context to a generative language model for answer generation. The generated answer is accompanied by source references to ensure transparency. Additionally, an integrated fact verification module evaluates the factual claims in the response, assigns a truthfulness score, and provides a confidence level with justification. The proposed system improves reliability by grounding responses in retrieved documents and validating them through automated evaluation. Implemented using Python, Streamlit, LangChain, FAISS, and Google Gemini APIs, the system demonstrates enhanced credibility compared to standalone generative models.

Arcot Naveen, Dr.D.William Albert, G Sreeramulu,"A Retrieval Based Integrated Fact Verification and Question Answering System" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(3): Page(581-587) May-June 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.