Cross-Lingual Text Augmentation: A Contrastive Learning Approach for Low-Resource Languages

Cross-lingual natural language understanding (NLU) is crucial for enabling multilingual applications, particularly in scenarios where labeled data and linguistic resources are limited. Addressing the challenge of knowledge transfer from high-resource to low-resource languages requires effective strategies to enhance model performance. This work explores approaches to improving cross-lingual representations and mitigating language gaps in low-resource settings.