Year: 2026
Type: article
Abstract: The rapid growth of digital documents across domains such as education, research, and enterprises has made efficient information retrieval increasingly challenging. Traditional keyword-based search systems often fail to provide accurate and context-aware results. DOCUMIND addresses this problem by introducing a conversational AI system that combines Retrieval-Augmented Generation (RAG) with a multi-agent architecture to enable intelligent interaction with documents. The system supports multiple file formats and transforms them into semantic embeddings stored in a vector database, allowing efficient and meaningful information retrieval. DOCUMIND employs specialized agents for document ingestion, retrieval, and response generation, working together to deliver precise, contextaware answers. By integrating advanced language models, semantic search, and conversational memory, the system improves accuracy, reduces irrelevant responses, and enhances user experience. This approach transforms static document repositories into interactive knowledge systems, making it highly useful for students, researchers, and professionals.
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7217735468