Picking BERT's Brain: Probing for Linguistic Dependencies in Contextualized Embeddings Using Representational Similarity Analysis

TL;DR

This paper uses Representational Similarity Analysis to explore how BERT's contextualized embeddings encode various linguistic dependencies, revealing that BERT captures these dependencies more than less salient controls.

Contribution

It introduces a novel RSA-based method to probe linguistic dependencies in BERT's embeddings, providing insights into what aspects of context are encoded.

Findings

BERT encodes subject-verb dependencies effectively.

Pronoun embeddings reflect antecedent relationships.

Full-sentence embeddings encode key head words.

Abstract

As the name implies, contextualized representations of language are typically motivated by their ability to encode context. Which aspects of context are captured by such representations? We introduce an approach to address this question using Representational Similarity Analysis (RSA). As case studies, we investigate the degree to which a verb embedding encodes the verb's subject, a pronoun embedding encodes the pronoun's antecedent, and a full-sentence representation encodes the sentence's head word (as determined by a dependency parse). In all cases, we show that BERT's contextualized embeddings reflect the linguistic dependency being studied, and that BERT encodes these dependencies to a greater degree than it encodes less linguistically-salient controls. These results demonstrate the ability of our approach to adjudicate between hypotheses about which aspects of context are encoded in representations of language.