Therapeutics Initiative Methods Speaker Series — Elicit: Applying language models to evidence synthesis

This talk introduces Elicit, an AI research assistant inspired by the systematic review process. The talk explores how Elicit uses language models and other machine learning methods to enable unique search and text extraction features. It also discusses the challenges of working with such new technology, notable failure modes, and insights from the current user base of researchers.

“Elicit can find relevant papers without perfect keyword match, summarize takeaways from the paper specific to your question, and extract key information from the papers.” – from Elicit.org website.

ABOUT THE SPEAKER:
Jungwon Byun is the COO and co-founder of Ought, a machine learning research lab. Ought builds Elicit, an AI research assistant using language models to automate parts of literature review. Elicit helps researchers find relevant papers, extract key information, and summarize them. Prior to Ought, Jungwon was the Head of Growth at Upstart, a fintech company using machine learning for consumer credit. She was also a management consultant and graduated cum laude from Yale with a B.A. in economics.

ABOUT THE THERAPEUTICS INITIATIVE METHODS SPEAKER SERIES:
The TI Methods Speaker Series are offered free of charge and everyone is welcome. The event is held at noon on the last Wednesday of each month. During the COVID-19 pandemic, while physical distancing measures are in effect, the TI Methods Speaker Series are offered via video-conference. The presentations are recorded and the video recordings are posted online.

Learn more on the event site.
https://www.ti.ubc.ca/2022/07/11/elicit/