Learning to Transform, Combine, and Reason in Open-Domain Question Answering
Users seek direct answers to complex questions from large open domain knowledge sources like the Web. In this paper, we propose a deep learning model based on the Transformer architecture that is able to efficiently operate over a larger set of candidate documents by effectively combining the evidence from these documents during multiple steps of reasoning, while it is robust against noise from low-ranked non-relevant documents included in the set.
Jaap Kamps is an associate professor of information retrieval at the University of Amsterdam, and a leading member of it’s world-class AI/NLP group. He mostly focused on text understanding for search and large scale information access. He has published over 400 papers in all major conferences and journals, see: