More about HKUST
Recent Progress on Machine Reading Comprehension
Speaker: Yiming CUI Director iFLYTEK Research Title: "Recent Progress on Machine Reading Comprehension" Date: Monday, 13 November 2017 Time: 4:00pm to 5:00pm Venue: Lecture Theater F (near lift 25/26), HKUST Abstract: Machine Reading Comprehension (MRC) has become a hot topic in recent natural language processing field, which aims to read and comprehend a given article or context and answer the questions based on it, requiring better understanding of natural languages and the ability to do reasoning over various clues. In this talk, I will give an introduction to reading comprehension task as well as its related datasets and technical approaches. Also, I will give a detailed introduction on our works including several competing neural network models, such as Attention-over-Attention Reader etc. At the end of this talk, I will envision future prospects in MRC and illustrate potential challenges in the future studies. ******************* Biography: Yiming Cui is a research director at iFLYTEK Research, China. He received his M.S. and B.S degree from Harbin Institute of Technology (HIT), majoring computer science. His main research interests include: Machine Reading Comprehension, Machine Translation and Natural Language Processing. He participated in several machine translation competitions, such as IWSLT 2012, IWSLT 2014, NIST OpenMT 15 and got several first prizes. In 2017, he leaded his team in Stanford Question Answering Dataset (SQuAD) Evaluation and got the first place among various competing teams. He also has publications in the top NLP conferences, such as ACL, COLING and NAACL and is reviewing papers for several conferences and journal, such as ACL, JCSL etc.