Modular Search Framework for Military Developers

Military developers often face unique challenges when searching for information due to the restrictive and specialized environments in which they operate. In recent years, Large Language Models (LLMs) have demonstrated exceptional capabilities in generating coherent, human-like text and answering complex queries across a range of natural language tasks. A modular architecture is ideal, where core LLM capabilities (e.g., code understanding, summarization, and retrieval) operate independently of the specific search engine. We propose a modular, adaptable information retrieval framework tailored for military use, which integrates LLMs as a core component and we developed a prototype based on our proposed framework and conducted a preliminary evaluation using a curated dataset. Our prototype achieved a recall of 95.94%. This modular and adaptable approach underscores the importance of integrating advanced information retrieval techniques in military contexts, paving the way for secure, efficient, and context-aware development processes. This paper was originally presented at the NATO Science and Technology Organization Symposium (ICMCIS) organized by the Information Systems Technology (IST) Scientific and Technical Committee, IST-209-RSY - the ICMCIS, held in Oeiras, Portugal, 13–14 May 2025.