Is neural language acquisition similar to natural? A chronological probing study

Abstract

The probing methodology allows one to obtain a partial representation of linguistic phenomena stored in the inner layers of the neural network, using external classifiers and statistical analysis. Pre-trained transformer-based language models are widely used both for natural language understanding (NLU) and natural language generation (NLG) tasks making them most commonly used for downstream applications. However, little analysis was carried out, whether the models were pre-trained enough or contained knowledge correlated with linguistic theory. We are presenting the chronological probing study of transformer English models such as MultiBERT and T5. We sequentially compare the information about the language learned by the models in the process of training on corpora. The results show that 1) linguistic information is acquired in the early stages of training 2) both language models demonstrate capabilities to capture various features from various levels of language, including morphology, syntax, and even discourse, while they also can inconsistently fail on tasks that are perceived as easy. We also introduce the open-source framework for chronological probing research, compatible with other transformer-based models. https://github.com/EkaterinaVoloshina/chronological_probing

arXiv commentPublished in proceedings of Dialogue-2022 "Computational Linguistics and Intellectual Technologies"

References in corpus (5)

• The Pile: An 800GB Dataset of Diverse Text for Language Modeling
• Well-Read Students Learn Better: On the Importance of Pre-training Compact Models
• SentEval: An Evaluation Toolkit for Universal Sentence Representations
• Evaluating Layers of Representation in Neural Machine Translation on Part-of-Speech and Semantic Tagging Tasks
• Understanding Learning Dynamics Of Language Models with SVCCA