Improving Polish to English Neural Machine Translation with Transfer Learning: Effects of Data Volume and Language Similarity
This work addresses improving translation quality for low-resource language pairs like Polish-English, but it is incremental as it builds on existing transfer learning methods.
The paper investigates how data volume and language similarity affect transfer learning for Polish-English neural machine translation, finding that combining related languages with larger data volumes yields better performance than either approach alone.
This paper investigates the impact of data volume and the use of similar languages on transfer learning in a machine translation task. We find out that having more data generally leads to better performance, as it allows the model to learn more patterns and generalizations from the data. However, related languages can also be particularly effective when there is limited data available for a specific language pair, as the model can leverage the similarities between the languages to improve performance. To demonstrate, we fine-tune mBART model for a Polish-English translation task using the OPUS-100 dataset. We evaluate the performance of the model under various transfer learning configurations, including different transfer source languages and different shot levels for Polish, and report the results. Our experiments show that a combination of related languages and larger amounts of data outperforms the model trained on related languages or larger amounts of data alone. Additionally, we show the importance of related languages in zero-shot and few-shot configurations.