LLMs In Healthcare - Code Lab
Explore the cutting edge of LLMs in this hands-on session dedicated to advancing the use of large language models (LLMs) in healthcare. This event will focus on best practices in prompt engineering and evaluation techniques critical for building trustworthy and effective AI systems.
What You'll Learn:
Prompt Engineering Techniques: Develop and refine prompts to optimize LLM performance for medical applications.
Evaluation Best Practices: Learn the importance of systematic prompt evaluations in enhancing interpretability and aligning AI behavior with your business's needs.
Hands-On Case Study: Work through a recent research paper that evaluates the ability of LLMs to mimic diagnostic reasoning processes using specialized prompts. Reproduce the study's results and experiment with improvements using advanced prompt engineering strategies.
Why Attend?
This session offers a unique opportunity to bridge technical innovation with healthcare applications. You’ll gain practical skills in effectively using LLMs while understanding how evaluation techniques can ensure AI systems meet the rigorous standards of healthcare environments.
Before Attending:
Please read the paper that we will be working through here: https://www.nature.com/articles/s41746-024-01010-1. You can find the data that we will be working with here.
Please have a working Python environment (python 3.10>=) and a code editor where you can run Jupyter notebooks.
You should also have an OpenAI api key that you can use to call their models.