RNNs as psycholinguistic subjects: Syntactic state and grammatical dependency

Abstract

Recurrent neural networks (RNNs) are the state of the art in sequence modeling for natural language. However, it remains poorly understood what grammatical characteristics of natural language they implicitly learn and represent as a consequence of optimizing the language modeling objective. Here we deploy the methods of controlled psycholinguistic experimentation to shed light on to what extent RNN behavior reflects incremental syntactic state and grammatical dependency representations known to characterize human linguistic behavior. We broadly test two publicly available long short-term memory (LSTM) English sequence models, and learn and test a new Japanese LSTM. We demonstrate that these models represent and maintain incremental syntactic state, but that they do not always generalize in the same way as humans. Furthermore, none of our models learn the appropriate grammatical dependency configurations licensing reflexive pronouns or negative polarity items.

References in corpus (5)

• Sequence to Sequence Learning with Neural Networks
• Semantics derived automatically from language corpora contain human-like biases
• Character-Aware Neural Language Models
• Exploring the Limits of Language Modeling
• Exploring the Syntactic Abilities of RNNs with Multi-task Learning

Cited by in corpus (12)

• On the Predictive Power of Neural Language Models for Human Real-Time Comprehension Behavior
• Mechanisms for Handling Nested Dependencies in Neural-Network Language Models and Humans
• A Systematic Assessment of Syntactic Generalization in Neural Language Models
• Can RNNs learn Recursive Nested Subject-Verb Agreements?
• Studying the Inductive Biases of RNNs with Synthetic Variations of Natural Languages
• Probing the phonetic and phonological knowledge of tones in Mandarin TTS models
• Structural Supervision Improves Learning of Non-Local Grammatical Dependencies
• Representation of Constituents in Neural Language Models: Coordination Phrase as a Case Study
• Shaking Syntactic Trees on the Sesame Street: Multilingual Probing with Controllable Perturbations
• Do RNNs learn human-like abstract word order preferences?
• How much pretraining data do language models need to learn syntax?
• The Language Model Understood the Prompt was Ambiguous: Probing Syntactic Uncertainty Through Generation