About
I am a fifth year PhD student in Computer Science at UMass Amherst advised by Mohit Iyyer who is now a professor at the University of Maryland.
My current research interests are in:
- long-context language models, particularly their evaluation and their applications to literary texts
- computer-use agents and how to improve their abilities to perform complex web-browsing tasks
Previously, I worked on machine translation, evaluation metrics, and detailed human evaluation + annotation collection.
Before coming to UMass, I was an undergraduate at Rutgers University where I majored in Math, English, and Computer Science. Prior to transferring to Rutgers, I spent one semester at the University of Edinburgh.
I've participated in 4 NSF REUs and done research in cool places like Hong Kong and the American Museum of Natural History in NYC.
Publications
BEARCUBS: A Benchmark for Computer-Using Web Agents
Yixiao Song, Katherine Thai, Chau Minh Pham, Yapei Chang, Mazin Nadaf, and Mohit Iyyer
One Thousand and One Pairs: A "Novel" Challenge for Long-Context Language Models
Exploring Document-Level Literary Machine Translation with Parallel Paragraphs from World Literature
Katherine Thai*, Marzena Karpinska*, Kalpesh Krishna, Moira Inghilleri,
John Wieting, and Mohit Iyyer
EMNLP 2022
🌾 DEMETR:
Diagnosing Evaluation Metrics for Translation
Marzena Karpinska, Nishant Raj, Katherine Thai, Yixiao Song,
Ankita Gupta, and Mohit Iyyer
EMNLP 2022
RELiC:
Retrieving Evidence for Literary Claims
ChapterBreak: A Challenge Dataset
for Long-Range Language Models
Teaching
I was a teaching assistant for CS685 Advanced Natural Language Processing (Fall 2021) at UMass Amherst.
As an undergraduate, I was a tutor/mentor for Rutgers University's Office for Diversity and Academic Success in the Sciences (ODASIS) for four years. I taught weekly Saturday classes to 20+ high school juniors and seniors in Algebra, English Composition, and SAT prep.
The layout and graphics for this site were designed by Sara Reed.