Thomas H. Chapin IV’s Post

https://lnkd.in/geT4Z4zp

A newspaper article from back when I was 16 years old

A recent article reports that GPT-4, when prompted with this new “MedPrompt” technique, outperforms even Med-PaLM 2 when asked a series of medical questions. https://lnkd.in/ghDMPrYw “With Medprompt, GPT-4 achieves state-of-the-art results on all nine of the benchmark datasets in the MultiMedQA suite. The method outperforms leading specialist models such as Med-PaLM 2 by a significant margin with an order of magnitude fewer calls to the model. Steering GPT-4 with Medprompt achieves a 27% reduction in error rate on the MedQA dataset over the best methods to date achieved with specialist models and surpasses a score of 90% for the first time.” Paper: https://lnkd.in/giwkiZej GitHub: https://lnkd.in/gEvpZVhg

Volunteering at the Foresight Vision Weekend event was exhausting, but rewarding! I’m very thankful that I got to be part of it! During the three day event, I got to meet and talk to numerous people who were interested in longevity and AI, including many people that I have been following and interacting with on social media for the past year. Kudos to Allison Duettmann and the rest of the team at Foresight Institute, and thanks to all of the speakers and sponsors for making this event possible!