A Simple and Effective Method to Improve Zero-Shot Cross-Lingual Transfer Learning
This work addresses the challenge of cross-lingual transfer for low-resource languages by reducing dependency on expensive parallel corpora, offering a novel method that improves alignment and performance in zero-shot settings.
The paper tackles the problem of zero-shot cross-lingual transfer learning by addressing the embedding gap between languages, proposing Embedding-Push, Attention-Pull, and Robust targets to improve transferability without relying on parallel data, resulting in significant performance gains on text classification tasks with mBERT and XLM-R models.
Existing zero-shot cross-lingual transfer methods rely on parallel corpora or bilingual dictionaries, which are expensive and impractical for low-resource languages. To disengage from these dependencies, researchers have explored training multilingual models on English-only resources and transferring them to low-resource languages. However, its effect is limited by the gap between embedding clusters of different languages. To address this issue, we propose Embedding-Push, Attention-Pull, and Robust targets to transfer English embeddings to virtual multilingual embeddings without semantic loss, thereby improving cross-lingual transferability. Experimental results on mBERT and XLM-R demonstrate that our method significantly outperforms previous works on the zero-shot cross-lingual text classification task and can obtain a better multilingual alignment.