Parametric Knowledge

Parametric knowledge refers to implicit knowledge embedded in the parameters (weights) of a neural network. It represents statistical or distributed information acquired during pretraining, rather than explicit facts stored in an external knowledge base. In RAG (Retrieval-Augmented Generation), it is contrasted with non-parametric knowledge, serving as the model’s internal “implicit value.”

In SimpleModeling, parametric knowledge forms the implicit layer of AI models, integrated with the external explicit layer (BoK: Body of Knowledge) to build a knowledge-cooperative foundation. This relationship can be organized as follows:

Through this structure, Pretrained Parametric Knowledge represents the initial pretrained state, which can be extended and updated via RAG and Assimilation processes.