Monday, October 30 • 4:40pm – 6:00pm EST

In recent years Large Language Models (LLMs) have emerged as a powerful technology capable of a range of tasks. This panel will discuss the potential of LLMs to democratize access to medical evidence by automatically simplifying medical texts. Specifically, in this session, attendees will learn about recent work investigating state-of-the-art LLMs for automatically simplifying such texts, and the current limitations to such methods. We will emphasize outstanding technical challenges (e.g., models “hallucinating” inaccurate content in simplified outputs), and connect these to downstream implications for public health.