Summary
AI is transforming the educational landscape. LLM technologies offer promising applications in synthetic persona generation for training and simulation across disciplines. The integration of synthetic actors and conversational AI agents into assignments enhances student engagement and streamlines instructor workload. Rather than watching training videos and submitting written responses, students and trainees interact with synthetic personas that mimic real-world scenarios. Instead of monitoring for AI misuse, instructors can focus on providing targeted, individualized feedback.
CAAI’s templated module for generating synthetic actors and conversational AI agents provides a scalable, interactive solution. Built on LLM Factory, we leverage training adapters for domain-specific customization, ensuring adaptability across different educational contexts. Human-in-the-loop feedback mechanisms further enhance model accuracy and relevance.
Associated projects include collaborations with the Center for Innovation in Population Health and the Office of Medical Education.
Datasets
For more information, please email ai@uky.edu.
Model
CAAI’s synthetic actor and conversational AI agent development is supported by LLM Factory, the base model used in these applications was Llama 3 8B.
A Low Rank Adapter (LoRA) was required to fine-tune the model to meet the needs of the social worker training application. Many vignettes contained sensitive areas of discussion, such as encounters with the juvenile justice system and drug use that created additional challenges related to safety and alignment.