Prompt Fine-Tuning of Generative Language Models On-Premises: Enhancing Knowledge Transfer Between Professors and Students
![default](/media/image/20/xlarge_2_1/default.jpg?nohash)
Prompt Fine-Tuning of Generative Language Models On-Premises: Enhancing Knowledge Transfer Between Professors and Students
The project aims to develop and test a new knowledge transfer and pedagogical tool to enhance the learning outcomes of university students. The novelty of the project relates to leveraging generative language models (GLMs) -- fine-tuned on-premises using reinforcement learning with human feedback (RLHF). The tool will allow professors to customize GLMs to different lectures and subjects while ensuring data protection and reducing bias in GLM outputs.