How Robust is Neural Machine Translation to Language Imbalance in Multilingual Tokenizer Training?
This work addresses a practical issue for developers of multilingual NLP systems by showing that tokenizer training is less sensitive to data imbalance than model training, though it is incremental in clarifying existing assumptions.
The study investigates how language imbalance in multilingual tokenizer training affects neural machine translation performance, finding that downstream translation is more robust to imbalance than expected, with UNK rate and closeness to character level serving as early warning indicators.
A multilingual tokenizer is a fundamental component of multilingual neural machine translation. It is trained from a multilingual corpus. Since a skewed data distribution is considered to be harmful, a sampling strategy is usually used to balance languages in the corpus. However, few works have systematically answered how language imbalance in tokenizer training affects downstream performance. In this work, we analyze how translation performance changes as the data ratios among languages vary in the tokenizer training corpus. We find that while relatively better performance is often observed when languages are more equally sampled, the downstream performance is more robust to language imbalance than we usually expected. Two features, UNK rate and closeness to the character level, can warn of poor downstream performance before performing the task. We also distinguish language sampling for tokenizer training from sampling for model training and show that the model is more sensitive to the latter.