| Following the recent development of LLMs it is suggested that these can now form the basis for Agentic AI. LLM-COGNICA is an approach to Agentic AI based on the integration of the Natural Language capabilities of LLMs with Argumentation-based Reasoning that is carried out on a Controlled Natural Language. It is a hybrid neural symbolic framework for developing decision making systems specified within Natural Language and executed via Explainable Argumentation. LLM-COGNICA offers the reliability of formal problem solving by allowing the human developer and problem solver to be in control of the agent system development process. The paper presents the LLM-COGNICA environment for the development of agent decision making, together with a systematic evaluation of the central element of mapping a decision policy in Natural Language to a corresponding policy in symbolic form. |
*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.