Low-resource Languages: A Review of Past Work and Future Challenges
arXiv:2006.07264
Abstract
A current problem in NLP is massaging and processing low-resource languages which lack useful training attributes such as supervised data, number of native speakers or experts, etc. This review paper concisely summarizes previous groundbreaking achievements made towards resolving this problem, and analyzes potential improvements in the context of the overall future research direction.
References in corpus (15)
- How multilingual is Multilingual BERT?
- Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation
- Low-Resource Named Entity Recognition with Cross-Lingual, Character-Level Neural Conditional Random Fields
- Universal Neural Machine Translation for Extremely Low Resource Languages
- Establishing Baselines for Text Classification in Low-Resource Languages
- Meta-Learning for Low-Resource Neural Machine Translation
- Low Resource Neural Machine Translation: A Benchmark for Five African Languages
- Investigating an approach for low resource language dataset creation, curation and classification: Setswana and Sepedi
- Language Graph Distillation for Low-Resource Machine Translation
- Model Transfer for Tagging Low-resource Languages using a Bilingual Dictionary
- Design Challenges in Low-resource Cross-lingual Entity Linking
- Towards Zero-resource Cross-lingual Entity Linking
- Low resource language dataset creation, curation and classification: Setswana and Sepedi -- Extended Abstract
- Unsupervised Morphological Paradigm Completion
- Improving Candidate Generation for Low-resource Cross-lingual Entity Linking