Greg Durrett
Associate Professor
Greg Durrett is an Assistant Professor in Computer Science. His current research covers a range of topics in statistical natural language processing, including coreference resolution, entity linking, document summarization, and question answering. Solving these problems lets computers access the information in unstructured text and transform this information in structured ways. Greg received his Ph.D. from UC Berkeley in 2016, where he was a part of the Berkeley NLP Group. He completed his B.S. at MIT in Computer Science and Mathematics in 2010.
Research
Research Areas:
Select Publications
. 2019. Understanding dataset design choices for multi-hop reasoning. NAACL.
. 2019. Learning to Denoise Distantly-Labeled Data for Entity Typing. NAACL.
. 2018. Spherical Latent Spaces for Stable Variational Autoencoders. EMNLP.
. 2018. Picking Apart Story Salads . EMNLP.
. 2014. A Joint Model for Entity Analysis: Coreference, Typing, and Linking. TACL.
Awards & Honors
- 2018 - NSF Small: Applying discrete reasoning steps in solving natural language processing tasks
- 2018 - NSF Medium: Computer-Aided Programming for Data Science
- 2017 - Bloomberg Data Science Research Grant
- 2014 - Facebook Fellowship in Natural Language Processing
- 2014 - Chair’s Special Award for Service
- 2013 - Best Paper Finalist, EMNLP
- 2012 - Outstanding Graduate Student Instructor Award
- 2010 - National Science Foundation Graduate Research Fellowship