Multilingual Syntax-aware Language Modeling through Dependency Tree Conversion
This work addresses the challenge of incorporating syntactic biases into multilingual language models, which is incremental as it builds on existing conversion methods and focuses on domain-specific improvements for syntax-aware modeling.
The study tackled the problem of extending syntax-aware language models to multilingual settings by evaluating dependency-to-constituency conversion methods, finding that the best model achieved a 19% increase in accuracy over the worst choice across five languages.
Incorporating stronger syntactic biases into neural language models (LMs) is a long-standing goal, but research in this area often focuses on modeling English text, where constituent treebanks are readily available. Extending constituent tree-based LMs to the multilingual setting, where dependency treebanks are more common, is possible via dependency-to-constituency conversion methods. However, this raises the question of which tree formats are best for learning the model, and for which languages. We investigate this question by training recurrent neural network grammars (RNNGs) using various conversion methods, and evaluating them empirically in a multilingual setting. We examine the effect on LM performance across nine conversion methods and five languages through seven types of syntactic tests. On average, the performance of our best model represents a 19 \% increase in accuracy over the worst choice across all languages. Our best model shows the advantage over sequential/overparameterized LMs, suggesting the positive effect of syntax injection in a multilingual setting. Our experiments highlight the importance of choosing the right tree formalism, and provide insights into making an informed decision.