Dieuwke Hupkes

Dieuwke's undergrad topic of study was (astro)physics, after which she transitioned to AI and natural language processing in during my MSc in Logic. Dieuwke holds a PhD in computational linguistic. In her research, she studies neural models of natural language processing, with a specific focus on interpretability, compositionality and the extent to which they can represent grammar and hierarchy.

Dieuwke's Publications

June 02, 2019

NLP

SPEECH & AUDIO

The emergence of number and syntax units in LSTM language models

Recent work has shown that LSTMs trained on a generic language modeling objective capture syntax-sensitive generalizations such as long-distance number agreement. We have however no mechanistic understanding of how they accomplish this remarkable feat.

Yair Lakretz, Germán Kruszewski, Theo Desbordes, Dieuwke Hupkes, Stanislas Dehaene, Marco Baroni

June 02, 2019