Approach to Physics Education Using Local AI with RAG for Open Educational Materials Generation

Presentation Type

Presentation

Start Date

8-4-2026 3:30 PM

End Date

8-4-2026 4:00 PM

Description

The new tool that starts to enter all parts of our life is AI – and it enters education as well. There are two standing large concerns with using AI for education – the safety of students’ data and the AI missing specific knowledge about the given class. The approach of using the Retrieval Augmented Generation (RAG) provides the user data to the locally run LLM model (using Ollama framework, a free and open-source tool that allows you to run large language models locally on your system) as a context for the generation of the OER materials. As this AI model of user’s choice is run fully locally, no data is transmitted and it can be used to do tasks with students’ data, such as summarization of evaluations, simple grading and creation of quiz materials using RAG.

Author Bios

Dr. Beznosko is the associate professor at Clayton State University. His interests are shared between the teaching and educational research, and the search for the new and exciting things in the vastness of the universe using ultra-high energy cosmic rays detection and analysis.

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

This document is currently not available here.

Share

COinS
 
Apr 8th, 3:30 PM Apr 8th, 4:00 PM

Approach to Physics Education Using Local AI with RAG for Open Educational Materials Generation

The new tool that starts to enter all parts of our life is AI – and it enters education as well. There are two standing large concerns with using AI for education – the safety of students’ data and the AI missing specific knowledge about the given class. The approach of using the Retrieval Augmented Generation (RAG) provides the user data to the locally run LLM model (using Ollama framework, a free and open-source tool that allows you to run large language models locally on your system) as a context for the generation of the OER materials. As this AI model of user’s choice is run fully locally, no data is transmitted and it can be used to do tasks with students’ data, such as summarization of evaluations, simple grading and creation of quiz materials using RAG.