The exponential growth of academic literature presents significant challenges for researchers attempting to find relevant information. Traditional keyword-based retrieval systems often fail to address issues such as synonyms, homonyms, and semantic nuances, leading to suboptimal search results. This paper introduces a novel system called IntelliSMART (Intelligent Semantic Machine-Assisted Research Tool), which leverages large language models (LLMs) and advanced semantic processing techniques to improve the retrieval of academic literature. Our approach integrates query rewriting, embedding generation, efficient indexing, and complex article retrieval mechanisms to provide highly accurate and contextually relevant results that align with the user’s intent. The IntelliSMART system features a user-friendly front end that facilitates intuitive query input, along with a robust back end for handling user queries, generating embeddings, indexing extensive collections of academic papers, and efficiently retrieving the most relevant documents. The proposed system shows significant improvements over conventional methods, highlighting its potential to transform the search experience in academic research.