Computational linguistics team wins international competition

A team of Georgetown students including Zhuoxuan Nymphea Ju (MS-CPLI 2026), Jingni Wu (MS-CPLI 2025), and Abhishek Purushothama (PhD-CS 2029) along with their mentor, Linguistics Associate Professor Amir Zeldes , submitted the top system for discourse relation classification in a recent international computational linguistics challenge. The team won first prize at the biannual Discourse Relation Parsing and Treebanking (DISRPT) competition. They will present their winning research virtually at the 2025 EMNLP, or Empirical Methods in Natural Language Processing, conference in Suzhou, China, on Nov. 9, 2025. Read extended coverage on their work in this article released by Georgetown University Graduate Studies Fellowship news here.

The DISRPT shared task  is a contest for models to identify often implicit relations between utterances such as understanding that “A is the cause of B”, or “B concedes a counterpoint to A” and so on. The team placed first among five entrants on the challenging task of relation classification. Read their paper about the system here.

Congratulations to the winning team! We wish you all the best in your virtual presentation in November!