Nishant Balepur

Hi! My name is Nishant and I’m a second-year Ph.D. student at the University of Maryland, where I am fortunate to be advised by Professors Jordan Boyd-Grayber and Rachel Rudinger. I am graciously supported by the NSF GRFP and a Cohere For AI Research Grant.

I am (super broadly) working on aligning and evaluating LLMs (or Muppets). My research can be grouped into 3 questions:

1. How can we make models more factual?
2. How can we guide models toward downstream user goals?
3. How can we better evaluate model weaknesses and capabilities?