Jay-Yoon Lee
Jay-Yoon Lee

Postdoctoral Researcher at
Information Extraction and Synthesis Laboratory,  
College of Information & Computer Sciences, UMass Amherst
Ph.D. in Computer Science
Computer Science Department,   School of Computer Scinece
Carnegie Mellon University

Email : jaylee @ iesl.cs.umass.edu
Here is my (slightly outdated) C.V.

Research Interests

The goal of my research is injecting knowledge/constraints into neural models, primarily for natural language processing (NLP) tasks. I am broadly interested in structured prediction, multi-task learning, logical reasoning, and better representation learning for the aforementioned topics. Currently I am exploring these topics as a postdoc research at IESL with Professor Andrew McCallum and previously with my advisor Jaime Carbonell during my Ph.D.

While Machine Learning is a great framework to learn a black-box model that best represents given training set, sometimes the model fails to learn simple rules that is necessary for safety or logical reasoning. However, if we step back and think, not all the aspects of the model need to be learned from the data itself as some constraints of the model are readily available to us. In this line of thought, I am interested in ‘how' current machine learning framework can incorporate knowledge that is in the form of constraints.



List of papers published/submitted to peer reviewed conferences/journals. Asterix (*) denotes joint first author, i.e., equal contribution.

Awards and Honors


Course Highlights at CMU

Work Experiences