PRESENTATION Bridging MDE and LLM-Based Agent Frameworks for Multi-Agent Systems

Key Points

• The aim is to explore how Model-Driven Engineering (MDE) can be integrated with LLM-based agent frameworks to enhance multi-agent systems.
• Conducted a quasi-systematic review of existing literature.
• Developed a metamodel to illustrate the integration of MDE and LLM frameworks.
• Presented findings at an international conference.
• Identified key interactions between MDE and LLM frameworks.
• Proposed a novel metamodel that facilitates integration.
• Highlighted potential improvements in multi-agent system development efficiency.