Towards Linguistically Generalizable NLP Systems: A Workshop and Shared Task
It addresses the problem of limited generalizability in NLP systems for researchers, but is incremental as it focuses on summarizing an existing workshop and task.
The paper summarizes a workshop and shared task aimed at testing the generalizability of NLP systems beyond their training data, highlighting results and lessons learned from the event.
This paper presents a summary of the first Workshop on Building Linguistically Generalizable Natural Language Processing Systems, and the associated Build It Break It, The Language Edition shared task. The goal of this workshop was to bring together researchers in NLP and linguistics with a shared task aimed at testing the generalizability of NLP systems beyond the distributions of their training data. We describe the motivation, setup, and participation of the shared task, provide discussion of some highlighted results, and discuss lessons learned.