Yale NLP/LLM Interest Group Talks

Join us for an upcoming NLP/LLM Interest Group Talk by Dr. Anran Li from Yale BIDS, who will discuss privacy and memorization challenges in Medical Large Language Models (MedLLMs). Dr. Li's research highlights how MedLLMs can inadvertently memorize and reveal sensitive patient information, posing privacy risks. The talk will cover methods for detecting and assessing memorization, factors influencing it, and strategies for mitigating these risks. This session is ideal for anyone interested in the secure and ethical development of AI in healthcare.