🤖 AI Summary
Traditional agent-centric architectures struggle to ensure semantic consistency, interpretability, and long-term stability in structured domains such as enterprises and institutional systems. This work proposes the World-centered Multi-Agent System (WMAS), a novel architectural paradigm that centers on the “world” as the primary modeling unit. By constructing a shared and explicit representation of the world, WMAS enables multi-agent learning and coordination grounded in a unified semantic model. Integrating semantic modeling, ontology engineering, and multi-agent coordination techniques, the approach structures world representations along dimensions such as ontological explicitness and normativity, thereby supporting global consistency and verifiable agent behaviors. The effectiveness of WMAS is demonstrated through the Ontobox platform, which validates its capacity to enhance semantic consistency, interpretability, and long-term system stability.
📝 Abstract
We introduce world-centered multi-agent systems (WMAS) as an alternative to traditional agent-centered architectures, arguing that structured domains such as enterprises and institutional systems require a shared, explicit world representation to ensure semantic consistency, explainability, and long-term stability. We classify worlds along dimensions including ontological explicitness, normativity, etc. In WMAS, learning and coordination operate over a shared world model rather than isolated agent-local representations, enabling global consistency and verifiable system behavior. We propose semantic models as a mathematical formalism for representing such worlds. Finally, we present the Ontobox platform as a realization of WMAS.