Using Priming to Uncover the Organization of Syntactic Representations in Neural Language Models
Year: 2019
Type: conference-paper
Abstract: Neural language models (LMs) perform well on tasks that require sensitivity to syntactic structure.Drawing on the syntactic priming paradigm from psycholinguistics, we propose a novel technique to analyze the representations that enable such success.By establishing a gradient similarity metric between structures, this technique allows us to reconstruct the organization of the LMs' syntactic representational space.We use this technique to demonstrate that LSTM LMs' representations of different types of sentences with relative clauses are organized hierarchically in a linguistically interpretable manner, suggesting that the LMs track abstract properties of the sentence.
DOI: 10.18653/v1/k19-1007
Institutions: Johns Hopkins University, Cornell University
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: hybrid
OpenAlex ID: W2976040964
