Minnesota Natural Language Processing Seminar Series: Natural Language Inference with Reasoning and Knowledge

The Minnesota Natural Language Processing (NLP) Seminar is a venue for faculty, postdocs, students, and anyone else interested in theoretical, computational, and human-centric aspects of natural language processing to exchange ideas and foster collaboration. The talks are every other Friday from 12 p.m. - 1 p.m. during the Fall 2021 semester.

This week's speaker, Yohan Jo (Amazon Alexa AI), will be giving a talk titled "Natural Language Inference with Reasoning and Knowledge."

Abstract

Our society is overloaded with information, and people are getting more passive and dependent on information delivered by technologies. In order to build machine learning models that support human decision-making in a truthful way, we need a better understanding of how humans reason and how to incorporate human-like reasoning and knowledge into computational models to process information. In this talk, I'm going to focus on the problem of natural language inference and present my recent work from an argumentation perspective. I will be presenting approaches that incorporate logical mechanisms and knowledge graphs, followed by a practical application of counterevidence retrieval system.

Biography

Yohan Jo has worked on natural language processing for various aspects of dialogue in domains like argumentation, education, and clinical notes. His interests lie in modeling human reasoning and emotion, and his Ph.D. thesis focuses on computational analysis and generation of argumentation. Yohan recently received a Ph.D. degree from Carnegie Mellon University and joined Amazon for the Alexa AI – Natural Understanding team.