ºÚÁϸ£ÀûÍø

Leveraging a Large Language Model (LLM) for personalized learning in education is promising, yet cloud-based LLMs pose risks around data security and privacy. To address these concerns, a locally stored Small Language Model (SLM) utilizing Retrieval-Augmented Generation (RAG) methods to support computing students’ learning was developed and deployed by Michael Liut, assistant professor at the University of Toronto.

His previous work has demonstrated that SLMs can match or surpass popular LLMs in handling conversational data while evaluating data privacy, scalability, and feasibility of local deployments. Additionally, during this session, he will discuss novel techniques where preliminary results appear promising in not just active conversation, but in summarization of personalized feedback. Furthermore, he will provide some insights towards recommended guidance strategies for learning in this context.

This event is open to teachers and support staff; registration is free and can be done . Free lunch and drinks will be available to those attending in person.

For inquiries, email academylearningteaching@tue.nl.