Optimizing Deep Transformers for Chinese-Thai Low-Resource Translation
This addresses machine translation for a specific low-resource language pair, representing an incremental improvement.
The paper tackled Chinese-Thai low-resource machine translation by optimizing deep Transformer models, achieving state-of-the-art performance in constrained evaluation.
In this paper, we study the use of deep Transformer translation model for the CCMT 2022 Chinese-Thai low-resource machine translation task. We first explore the experiment settings (including the number of BPE merge operations, dropout probability, embedding size, etc.) for the low-resource scenario with the 6-layer Transformer. Considering that increasing the number of layers also increases the regularization on new model parameters (dropout modules are also introduced when using more layers), we adopt the highest performance setting but increase the depth of the Transformer to 24 layers to obtain improved translation quality. Our work obtains the SOTA performance in the Chinese-to-Thai translation in the constrained evaluation.