Practical Comparable Data Collection for Low-Resource Languages via Images
Preprint
Abstract
We propose a method of curating high-quality comparable training data for low-resource languages with monolingual annotators. Our method involves using a carefully selected set of images as a pivot between the source and target languages by getting captions for such images in both languages independently. Human evaluations on the English-Hindi comparable corpora created with our method show that 81.1% of the pairs are acceptable translations, and only 2.47% of the pairs are not translations at all. We further establish the potential of the dataset collected through our approach by experimenting on two downstream tasks - machine translation and dictionary extraction. All code and data are available at https://github.com/madaan/PML4DC-Comparable-Data-Collection.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Universal Phone Recognition with a Multilingual Allophone System2020
- Massively Multilingual Adversarial Speech Recognition2019
- Learning Multilingual Word Embeddings Using Image-Text Data2019
- Multimodal Pivots for Image Caption Translation2016
- Multi30K: Multilingual English-German Image Descriptions2016
- Neural Machine Translation of Rare Words with Subword Units2016
- Framing Image Description as a Ranking Task: Data, Models and Evaluation MetricsJournal of Artificial Intelligence Research, 2013