At 9:00 AM on January 17, 2025, Unlimited Research Group of AI (URA) had the honor of hosting Dr. Trang Vu from Monash University. During this engaging talk, Dr. Trang shared valuable insights into continual learning of Large Language Models (LLMs) and advanced methods related to the field.
Dr. Trang began by explaining the fundamental concepts of Large Language Models, including key stages such as pretraining, fine-tuning, and model alignment. She then delved into how LLMs can be aligned with the principles of Helpful, Honest, Harmless, utilizing techniques such as Reinforcement Learning with Human Feedback (RLHF). Beyond the technical aspects, Dr. Trang also shared her personal experiences as a PhD candidate at Monash University, providing practical advice on overcoming challenges in research. She offered guidance to URA members on pursuing study-abroad opportunities and building careers in international research environments.
The session left URA members with not only a deeper understanding of LLMs and continual learning but also inspiration to explore and innovate further. URA sincerely thanked Dr. Trang Vu for taking the time to visit and share such valuable knowledge!