Henderson, James2021-04-132021-04-132021-04-13202010.18653/v1/2020.acl-main.561https://infoscience.epfl.ch/handle/20.500.14299/177251In this paper, we trace the history of neural networks applied to natural language understanding tasks, and identify key contributions which the nature of language has made to the development of neural network architectures. We focus on the importance of variable binding and its instantiation in attention-based models, and argue that Transformer is not a sequence model but an induced-structure model. This perspective leads to predictions of the challenges facing research in deep learning architectures for natural language understanding.text::conference output::conference proceedings::conference paper