Causes and Cures for Interference in Multilingual Translation
This work addresses interference issues in multilingual translation models, which is an incremental improvement for machine learning practitioners.
The study investigated the causes of interference in multilingual machine translation, finding that model size, data size, and language pair proportions are key factors, with interference largely alleviated in standard transformers under one billion parameters and tuning sampling temperature improving overall performance.
Multilingual machine translation models can benefit from synergy between different language pairs, but also suffer from interference. While there is a growing number of sophisticated methods that aim to eliminate interference, our understanding of interference as a phenomenon is still limited. This work identifies the main factors that contribute to interference in multilingual machine translation. Through systematic experimentation, we find that interference (or synergy) are primarily determined by model size, data size, and the proportion of each language pair within the total dataset. We observe that substantial interference occurs mainly when the model is very small with respect to the available training data, and that using standard transformer configurations with less than one billion parameters largely alleviates interference and promotes synergy. Moreover, we show that tuning the sampling temperature to control the proportion of each language pair in the data is key to balancing the amount of interference between low and high resource language pairs effectively, and can lead to superior performance overall.