Robert Frank

Finding Syntactic Representations in Neural Stacks

2019 · arXiv (Cornell University)

Neural network architectures have been augmented with differentiable stacks in order to introduce a bias toward learning hierarchy-sensitive regularities. It has, however, proven difficult to assess the degree to which such a bias is effective, …

False perspectives on human language: Why statistics needs linguistics

2023 · Frontiers in Language Sciences

A sharp tension exists about the nature of human language between two opposite parties: those who believe that statistical surface distributions, in particular using measures like surprisal, provide a better understanding of language processing, vs. …

Open Sesame: Getting inside BERT’s Linguistic Knowledge

2019

How and to what extent does BERT encode syntactically-sensitive hierarchical information or positionally-sensitive linear information? Recent work has shown that contextual representations like BERT perform well on tasks that require sensitivity to linguistic structure. We …