Summary of Cantonmt: Cantonese to English Nmt Platform with Fine-tuned Models Using Synthetic Back-translation Data, by Kung Yin Hong et al.
CantonMT: Cantonese to English NMT Platform with Fine-Tuned Models Using Synthetic Back-Translation Data
by Kung Yin Hong, Lifeng Han, Riza Batista-Navarro, Goran Nenadic
First submitted to arxiv on: 17 Mar 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary Neural Machine Translation (NMT) for low-resource languages, such as Cantonese-to-English, is a challenging task in NLP. This paper proposes a standard data augmentation methodology by back-translation to improve the performance of NMT models. The authors fine-tune their models using limited real data and synthetic data generated through back-translation from OpusMT, NLLB, and mBART. Automatic evaluation metrics like lexical-based and embedding-based are used to assess model performance. Additionally, a user-friendly interface is created for the Cantonese-to-English MT research project, named CantonMT, with an open-source toolkit allowing researchers to add more models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps machines understand Cantonese-to-English translation better. It’s hard because there isn’t much data available. The researchers use a special trick called back-translation to make the data bigger. They test different ways of translating and compare results using special metrics. To help others, they create an easy-to-use platform called CantonMT with tools for adding more models. |
Keywords
» Artificial intelligence » Data augmentation » Embedding » Nlp » Synthetic data » Translation
