Bilingual Alignment Pre-Training for Zero-Shot Cross-Lingual Transfer
This work addresses the challenge of zero-shot cross-lingual transfer for multilingual NLP tasks, representing an incremental improvement over existing methods.
The paper tackled the problem of poor alignment in multilingual pre-trained models like mBERT by proposing a pre-training task called Word-Exchange Aligning Model (WEAM) that uses statistical alignment information to guide cross-lingual word prediction, resulting in significant improvements in zero-shot performance on tasks like MLQA and XNLI.
Multilingual pre-trained models have achieved remarkable performance on cross-lingual transfer learning. Some multilingual models such as mBERT, have been pre-trained on unlabeled corpora, therefore the embeddings of different languages in the models may not be aligned very well. In this paper, we aim to improve the zero-shot cross-lingual transfer performance by proposing a pre-training task named Word-Exchange Aligning Model (WEAM), which uses the statistical alignment information as the prior knowledge to guide cross-lingual word prediction. We evaluate our model on multilingual machine reading comprehension task MLQA and natural language interface task XNLI. The results show that WEAM can significantly improve the zero-shot performance.