Research topic
Semantic understanding for AI apps
Bayon studies how meaning, context, roles, and intent can become part of the structure of an application, not just the content of a prompt.
Meaning as product material
Most software treats language as text and AI as response generation. Bayon treats meaning as a product material: something that can be modeled, refined, remembered, and used to shape how an application behaves.
Context before completion
An AI-enabled app should know what role it is playing, what the user is trying to do, what information matters, and what should remain under human control. Semantic understanding gives that context structure.
From prompt to system
Prompting is useful, but serious products need repeatable systems: entities, roles, memories, rules, workflows, interfaces, and feedback loops that make meaning operational.
Applied in the portfolio
Semantic understanding shows up across Bayon projects: cognition engines that model conditions, knowledge platforms that structure information, and assistants that need durable context.