While models must perform reasoning to understand human language, this is mostly required for downstream applications and not as a standalone skill. The challenges of end-to-end reasoning in downstream applications, with a specific emphasis on end-to-end arithmetic reasoning, is the topic of "Challenges of End-to-End Reasoning in NLP", a talk by Nafise Sadat Moosavi of the University of Sheffield. The talk is part of OFAI's 2023 Fall Lecture Series.
Members of the public are cordially invited to attend the talk via Zoom on Wednesday, 11 October 2023 at 18:30 CEST (UTC+2):
URL: https://us06web.zoom.us/j/84282442460?pwd=NHVhQnJXOVdZTWtNcWNRQllaQWFnQT09
Meeting ID: 842 8244 2460
Passcode: 678868
You can add this event to your calendar.
Talk abstract: To understand human language, language models have to perform various reasoning skills, e.g., logical reasoning, commonsense reasoning, temporal reasoning, etc. There are multiple datasets for directly evaluating each of these reasoning skills. However, these reasoning skills are mostly required for downstream applications and not as standalone skills. For instance, a model may need to perform arithmetic reasoning for answering a question or to correctly summarize a table. However, it is not clear whether a model that performs well on a dataset that is designed to evaluate arithmetic reasoning would also improve the results on a QA dataset that requires arithmetic reasoning. As a result, we should pay special attention to developing end-to-end models for downstream applications that are also capable of performing various reasoning skills. This presentation focuses on the challenges of end-to-end reasoning in downstream applications, with a specific emphasis on end-to-end arithmetic reasoning.
Speaker biography: Dr. Nafise Sadat Moosavi is a Lecturer in Natural Language Processing at the Computer Science Department of the University of Sheffield. Before joining the University of Sheffield, she was a postdoctoral researcher at the Technical University of Darmstadt. She works on the limitation of language models to improve their fairness, reasoning, robustness, and efficiency. She co-founded and co-organizes SustaiNLP workshops and regularly serves as a senior area chair and area chair at *ACL conferences.
