Summary of Predicting Machine Translation Performance on Low-resource Languages: the Role Of Domain Similarity, by Eric Khiu et al.
Predicting Machine Translation Performance on Low-Resource Languages: The Role of Domain Similarity
by Eric Khiu, Hasti Toossi, David Anugraha, Jinyu Liu, Jiaxu Li, Juan Armando Parra Flores, Leandro Acros Roman, A. Seza Doğruöz, En-Shiun Annie Lee
First submitted to arxiv on: 4 Feb 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
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 The abstract discusses the challenges of fine-tuning and testing multilingual large language models for low-resource languages (LRLs). Researchers have previously predicted NLP task performance using machine learning methods, but mostly focused on high-resource languages. This study investigates three factors affecting model performance in LRLs: fine-tuning corpus size, domain similarity between training and testing corpora, and language similarity between source and target languages. Using classical regression models, the authors assess how these factors impact Machine Translation model performance. The results show that domain similarity has the most significant impact on predicting model performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper investigates how to predict the performance of machine translation models for low-resource languages. It looks at three things: how big the training data is, how similar the training and testing data are, and how similar the language being translated from is to the language being translated to. The researchers use simple math formulas to see how these factors affect model performance. They find that the similarity between the data used for training and testing has the biggest impact on how well the model does. |
Keywords
* Artificial intelligence * Fine tuning * Machine learning * Nlp * Regression * Translation
