I received my PhD in computational linguistics in the University of Tokyo, followed by a postdoc in the University of Massachusetts focusing on machine learning for Natural Language Processing (NLP). Currently I hold a postdoctoral position in the center for Social Data Science at the University of Copenhagen. My main research area is analysis of pre-trained meaning representations used in the modern deep learning-based NLP systems. I have worked extensively on analysis of word embeddings and BERT, and their applications in such areas as sentiment analysis, question answering, and temporal relation extraction. Most recently I also work on the ethical and legal aspects of data used in NLP research.