Although knowledge is currently accessible in ways previously considered impossible, much of it remains out of reach of non-expert information seekers. While question answering (QA) is an intuitive and flexible interface to help users, current paradigms in QA rarely consider the fundamental roles of a dynamically changing world and diverse information seekers. In this talk, I will present our work to bring them into the picture. First, we integrate temporal and geographical contexts into QA tasks to situate answers for users in diverse contexts. Then I will discuss leveraging user feedback to continuously improve the model after its deployment. I will conclude with a study of discourse structure in long form answers, which can open doors to satisfy users' complex information needs.
Eunsol Choi is an assistant professor in the Computer Science department at the University of Texas at Austin and a visiting researcher at Google AI. Her research spans natural language processing and machine learning. She is particularly interested in interpreting and reasoning about text in dynamic real-world contexts. She received a Ph.D. in Computer Science and Engineering from University of Washington and B.A in mathematics and computer science at Cornell University. She is a recipient of Facebook research fellowship, Google faculty research award, and outstanding paper award at EMNLP 2021.