Citation :

Vidya Vishal Wadkar, "

Contextual Coherence in Conversational AI: Leveraging a Memory Agent

Abstract

This study presents a Memory Agent Framework specifically designed for engaging conversational AI while addressing context coherence and memory scalability. The design is based on a Model Context Protocol (MCP) to synchronise many conversational agents over four memory layers inspired by cognitive functions (Short-Term, Episodic, Semantic, and Procedural). The Python-based asynchronous orchestration helps quickly retrieve memories, using FAISS/Pinecone vector storage and a Neo4j knowledge graph to dynamically reason a conversation (like a human would use memories). When the Memory Agent Framework was evaluated in three different modes (baseline, fine-tuned, and embedding-enhanced), it showed a real performance advantage through the Memory Agent Framework compared with baseline evaluation mode. In the Short-Term Memory, each perplexity level decreased from 327.18 to 294.47; in the Episodic Memory, it decreased from 348.49 to 313.64; and in Semantic Memory, it decreased from 344.22 to 309.79. However, the semantic coherence improved by 28.5% in contextual reliability, reaching 0.5499. The fine-tuned models achieve BLEU and ROUGE-L scores above 0.5, indicating improved grammatical correctness and relevance. These results suggest that the Memory-driven paradigm improves multi-turn conversation comprehension, contextual fragmentation, and episodic interaction continuity and understanding. Overall, the Memory Agent architecture supports scalable, context-rich conversation systems that promote coherence and adaptive reasoning in real-world communication.

Keywords

Multi-Agent Conversational AI, Memory Agent Framework, Model Context Protocol, Layered Memory Architecture, Contextual Coherence.

References

[1] Jin Chen et al., “When Large Language Models Meet Personalisation: Perspectives of Challenges and Opportunities,” World Wide Web, vol. 27, no. 4, pp. 1-45, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[2] Dyu Ghosh, Dibya Ghosh, and Debi Prasad Ghosh, “Brain-MCP: A Fully Homomorphic Neuro-Symbolic Protocol for Governing System 3 Cognition and Restoring Semantic Scarcity,” Manuscript, 2025.
[Google Scholar]
[3] Lorenz Cuno Klopfenstein et al., “The Rise of Bots: A Survey of Conversational Interfaces, Patterns, and Paradigms,” Proceedings of the 2017 Conference on Designing Interactive Systems, pp. 555-565, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Sakhinana Sagar Srinivas, Vijay Sri Vaikunth, and Venkataramana Runkana, “Towards Automated Patent Workflows: AI-Orchestrated 24 Vidya Vishal Wadkar / IJCTT, 73(11), 17-25, 2025 Multi-Agent Framework for Intellectual Property Management and Analysis,” arXiv preprint arXiv:2409.19006, pp. 1-16, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Ugochukwu Francis Ikwuanusi et al., “AI-Powered Real-Time Emotion Recognition: Pioneering Solutions for User Interaction and Engagement,” International Journal of Engineering Research and Development, vol. 20, no. 11, pp. 1188-1205, 2024.
[Google Scholar] [Publisher Link]
[6] Iuliia Kotseruba, and John K. Tsotsos, “40 Years of Cognitive Architectures: Core Cognitive Abilities and Practical Applications,” Artificial Intelligence Review, vol. 53, no. 1, pp. 17-94, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Qinghua Zheng et al., “Machine Memory Intelligence: Inspired by Human Memory Mechanisms,” Engineering, pp. 1-12, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Raymond Ajax, Ayuns Luz, and Barnabas Bamigboye, “Natural Language Processing (NLP) in AI-Driven Memory Recall Systems,” pp. 1-23, 2025.
[Google Scholar]
[9] Ali Nawaz, and Amir Ahmad, “Conversational AI for Cognitive, Emotional, and Social Engagement of Elderly Persons: A Large Language Model-Based Framework,” International Joint Conference Artificial Intelligence, pp. 17-28, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Thanh-Hai Nguyen, Dinh-Lam Pham, and Kwanghoon Pio Kim, “Next process-activity prediction using Switch-Transformer: approach, visualisation, and performance evaluation,” Knowledge and Information Systems, pp. 1-28, 2025.
[CrossRef] [Google Scholar] [Publisher Link] 
[11] Rachel Katharine Sterken, and James Ravi Kirkpatrick, “Conversational Alignment with Artificial Intelligence in Context,” Philosophical Perspectives, vol. 38, no. 1, pp. 89-102, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Yuntao Wang et al., “Large Model-Based Agents: State-of-the-Art, Cooperation Paradigms, Security and Privacy, and Future Trends,” IEEE Communications Surveys & Tutorials, pp. 1-1, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Ranjan Sapkota, Konstantinos I. Roumeliotis, and Manoj Karkee, “AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges,” arXiv preprint arXiv:2505.10468, pp. 1-30, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Xinming Wang et al., “The Hitchhiker's Guide to Autonomous Research: A Survey of Scientific Agents,” Authorea Preprints, pp. 1-24, 2025.
[Google Scholar]
[15] Yuwei Wan et al., “Empowering LLMs by Hybrid Retrieval-Augmented Generation for Domain-Centric Q&A in Smart Manufacturing,” Advanced Engineering Informatics, vol. 65, pp. 1-16, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[16] José Antonio Heredia Álvaro, and Javier González Barreda, “An Advanced Retrieval-Augmented Generation System for Manufacturing Quality Control,” Advanced Engineering Informatics, vol. 64, pp. 1-15, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Nur Arifin Akbar et al., “RAG-Driven Memory Architectures in Conversational LLMs Literature Review with Insights into Emerging Agriculture Data Sharing,” IEEE Access, vol. 13, pp. 123855-123880, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Chidiebere Joshua et al., “Building Retrieval-Augmented Generation (RAG) for Internal Knowledge Bases,” pp. 1-32, 2024.
[Google Scholar]