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{183483,
        author = {Aneeqa Imtiyaz and Ms. Rupinder Kaur},
        title = {Enhancing Retrieval Augmented Generation with Conversational Memory: A Domain-Grounded Context-Aware Architecture},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {3},
        pages = {1805-1812},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=183483},
        abstract = {In the constantly evolving field of Natural Language Processing, Retrieval-Augmented Generation models have emerged as a powerful tool to enhance the contextual relevance of large language models. However, most RAG implementations lack long-term memory and session awareness, restricting their ability to hold meaningful multi-turn talks. This study proposes a database-driven, context-aware RAG architecture for improving conversational continuity by extracting top-N historical interactions from a MongoDB session database. These are integrated with document-based context obtained through vector search in ChromaDB. The final prompt, created with both document and session context, is sent to a Gemini 1.5 Pro model via API for generation. The suggested approach was tested across several sessions with six domain-specific papers. The results of evaluation showed high relevance of context and correct responses, especially in further requests. The proposed architecture can be seen as a bridge between traditional RAG systems and conversational agents that are capable of continuing sessions. It tries to introduce an efficient way for producing flexible responses.},
        keywords = {Context-awareness, Conversational AI, Large Language Model, Natural Language Processing, Retrieval-Augmented Generation, session memory.},
        month = {August},
        }