Taught a three hour lecture for professionals who want to understand more about the inner workings of LLMs

Published a new fact-checking benchmark, verifying claims about COVID-19 in the news using evidence from scientific articles

Built upon a popular gender debiasing method for word embeddings by broadening the method for names swapping to function on datasets containing uneven numbers of male and female names through the implementation of a many-to-many probabilistic matching algorithm that produces similar scores on debiasing evaluations to those from the original paper

Explored new training methods to simplify and improve COVID-19 misinformation detection, including leveraging new models, model architectures and datasets

Created a model to predict the future health of New York’s half a million street trees, using data collected from NYC street tree censuses and publicly available land use data

Collaborated on a method to analyze CMU Delphi’s abundant and disparate Covid-19 data as prospective leading indicators of case growth on the county level, culminating in a presentation at the June 2021 Symposium for Data Science and Statistics