Bringing Up a Bilingual BabyLM: Investigating Multilingual Language Acquisition Using Small-Scale Models

TL;DR

This study uses controlled language model training to simulate bilingual language acquisition, finding that different exposure regimes do not significantly hinder learning, and bilingual input is manageable for statistical learners.

Contribution

It introduces a method to simulate bilingual language acquisition with small-scale models, providing insights into how exposure regimes affect learning outcomes.

Findings

Bilingual models perform similarly to monolingual models in one language.

Bilingual models show strong performance in the second language.

Different exposure regimes do not significantly impact learning outcomes.

Abstract

Multilingualism is incredibly common around the world, leading to many important theoretical and practical questions about how children learn multiple languages at once. For example, does multilingual acquisition lead to delays in learning? Are there better and worse ways to structure multilingual input? Many correlational studies address these questions, but it is surprisingly difficult to get definitive answers because children cannot be randomly assigned to be multilingual and data are typically not matched between languages. We use language model training as a method for simulating a variety of highly controlled exposure conditions, and create matched 100M-word mono- and bilingual datasets using synthetic data and machine translation. We train GPT-2 models on monolingual and bilingual data organized to reflect a range of exposure regimes, and evaluate their performance on perplexity, grammaticality, and semantic knowledge. Across model scales and measures, bilingual models perform similarly to monolingual models in one language, but show strong performance in the second language as well. These results suggest that there are no strong differences between different bilingual exposure regimes, and that bilingual input poses no in-principle challenges for agnostic statistical learners.