PhD Dissertation Proposal by WANG Shuohang | Textual Sequence Matching for Question Answering

About The Talk

Question answering (QA) is one of the most important applications in natural language processing. With the explosive text data from the Internet, intelligently getting answers of questions will help humans more efficiently collect useful information. My research in this proposal mainly focuses on building vectorized representations for pairs of text sequences to enable better matching. The text sequences can be a question and a answer, so that the ground-truth answer that matches or entails the question would be identified based on the matching representations. Overall, our proposed models achieved state of the art on multiple public QA tasks including SQuAD, MSMARCO, WikiQA, MovieQA, InsuranceQA, and we will cover open-domain QA tasks, like SearchQA, TriviaQA and Quasar, in the future work.

Speaker Biography

WANG Shuohang is a PhD candidate at School of Information Systems, Singapore Management University advised by Associate Professor JIANG Jing and Associate Professor ZHENG Baihua. His research interests are in deep learning and reinforcement learning in Natural Language Processing which spans Question Answering, Machine Reading Comprehension, Text Entailment, etc..