Jessica Lin
My name is Jessica (Yi-Ju) Lin (林宜儒). I am an incoming Assistant Professor in the Department of Foreign Languages and Literatures at National Taiwan University. Before joining NTU, I worked as a Linguist at Google, where I contributed to Factuality dataset creation for model training and evaluation. I completed my PhD in Computational Linguistics at Georgetown University, where I was advised by Amir Zeldes and was a member of Corpling Lab and Computational Linguistics @ Georgetown (GUCL). Before coming to Georgetown, I recieved my M.A. in Linguistics from National Taiwan University and B.A. in Applied Chinese Language and Culture from National Taiwan Normal University. Check out my CV here.
Research Interests
My research develops computational and corpus-based approaches to discourse-level language understanding across genres. I am especially interested in how important information is introduced, maintained, and interpreted in longer texts, including questions such as which entities become central in a document, how readers and models identify them, and how their salience shapes prediction and summarization. I use these questions to study and evaluate large language models on discourse-level phenomena such as entity salience, summarization, hallucination, and discourse relations. More broadly, my work contributes to corpus linguistics through the creation and analysis of genre-diverse NLP datasets, and I am interested in extending these methods to literary texts and Digital Humanities.