Please give a brief, simplified overview of your research project.
Large Language Models (LLMs) are big machine learning models that can solve various natural language processing (NLP) tasks given only textual instructions. For example, you can instruct an LLM to classify movie reviews as positive or negative simply by giving a few examples of positive and negative reviews. I am studying ways we can better specify these textual instructions, known as prompts, to improve LLM performance on answering questions that are ambiguous, and can thus have multiple answers.
Describe the tasks you engage in as part of your work.
Most of my time is spent reading papers and planning what things to try, and then implementing those things in code and running experiments.
Describe what you thought college might be like before you came to UT. Did you consider research when thinking about college?
I definitely did consider research when thinking about college. I competed in science fairs in high school and developed an interest in scientific research. I was surprised when I came to UT about how easy it was to get involved in research and how many opportunities were available.
How did you get involved with your research project?
I took my professor’s natural language processing class last semester and became interested in the subject. Then after the semester ended, I decided to reach out about joining the lab.
Do you see your project connecting with your plans for your future?
I plan to do a Ph.D. in robotics. I think my current research will be quite valuable as there are many intersections between NLP and robotics, and NLP research has exposed me to a new way of thinking I believe will be very useful to me.
What is the most interesting or surprising thing you’ve gotten to do for this project?
Taking a robot home with me for winter break! The robot I’m running experiments on is small enough to transport easily, so I was able to continue my research over the break while I was away from campus.