Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{185943,
        author = {Gopichand Agnihotram and Joydeep Sarkar and Neha Maurya},
        title = {Enhancing Retrieval-Augmented Generation Systems with Hypothesis Testing},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {5},
        pages = {3497-3504},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=185943},
        abstract = {As large language models (LLMs) are increasingly embedded into knowledge-intensive applications through Retrieval-Augmented Generation (RAG), ensuring that both retrieval and response generation are meaningful and reliable becomes critical. Hypothesis testing, a foundational tool in inferential statistics, offers a principled framework to assess and control the reliability of retrieved contexts and generated outputs. This white paper explores how hypothesis testing can be integrated into RAG systems to improve contextual relevance, reduce hallucinations, and enhance the alignment between system outputs and source-grounded truth. We present a mathematically motivated foundation, practical formulations, and use-case driven illustrations for embedding statistical rigor into retrieval and generation.},
        keywords = {Retrieval-Augmented Generation (RAG), Hypothesis Testing, Large Language Models (LLMs), Hallucination Reduction, Natural Language Inference (NLI), Statistical Reliability},
        month = {October},
        }