- 11월 25일 정기 수요세미나 (온라인) 안내
- 컴퓨터과학과 홈페이지 관리자
- 게시글 내용
제목: Integration of Web Knowledge into Task-oriented Conversational Modeling
강사: 김석환 박사님 (Amazon Alexa AI)
일시: 2020년 11월 25일 (수요일) 오전 9시
Most prior work on task-oriented dialogue systems are restricted to a limited coverage of domain APIs, while users oftentimes have domain related requests that are not covered by the APIs. In this talk, we will discuss how to expand coverage of task-oriented dialogue systems by incorporating external unstructured knowledge sources with the following main tasks: knowledge-seeking turn detection, knowledge selection, and knowledge-grounded response generation. Recently, we’ve completed a public benchmark on those tasks as a main track of the Ninth Dialog System Technology Challenge (DSTC9). I will present the details of our data sets, baseline models and evaluation set-up as well as an overview of the challenge results.
Seokhwan Kim is a Senior Machine Learning Scientist at Amazon Alexa AI. Prior to joining Amazon, Seokhwan conducted work in natural language understanding and spoken dialog systems where he was an NLP Research Scientist at Adobe Research and a Research Scientist at the Institute for Infocomm Research (I2R) in Singapore. His PhD work focused on cross-lingual weakly-supervised language understanding at Pohang University of Science and Technology (POSTECH) in Korea, his advisor was Prof. Gary Geunbae Lee. He has authored 60 peer-reviewed papers in international journals and conferences with more than 900 citations. More recently, he has focused on knowledge-grounded conversional modeling and has contributed to the research community by co-organizing Dialog System Technology Challenges (DSTCs).