CoSDA-ML: Multi-Lingual Code-Switching Data Augmentation for Zero-Shot Cross-Lingual NLP
arXiv:2006.06402
Abstract
Multi-lingual contextualized embeddings, such as multilingual-BERT (mBERT), have shown success in a variety of zero-shot cross-lingual tasks. However, these models are limited by having inconsistent contextualized representations of subwords across different languages. Existing work addresses this issue by bilingual projection and fine-tuning technique. We propose a data augmentation framework to generate multi-lingual code-switching data to fine-tune mBERT, which encourages model to align representations from source and multiple target languages once by mixing their context information. Compared with the existing work, our method does not rely on bilingual sentences for training, and requires only one training process for multiple target languages. Experimental results on five tasks with 19 languages show that our method leads to significantly improved performances for all the tasks compared with mBERT.
Accepted at IJCAI2020. SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence), all rights reserved. http://static.ijcai.org/2020-accepted_papers.html
References in corpus (8)
- Cross-lingual Language Model Pretraining
- On the Cross-lingual Transferability of Monolingual Representations
- Unicoder: A Universal Language Encoder by Pre-training with Multiple Cross-lingual Tasks
- Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond
- Learning Bilingual Word Representations by Marginalizing Alignments
- Cross-Lingual BERT Transformation for Zero-Shot Dependency Parsing
- Unsupervised Multilingual Word Embeddings
- Dialog State Tracking with Reinforced Data Augmentation
Cited by in corpus (5)
- A Survey of Code-switched Speech and Language Processing
- M3P: Learning Universal Representations via Multitask Multilingual Multimodal Pre-training
- DICT-MLM: Improved Multilingual Pre-Training using Bilingual Dictionaries
- Cross-lingual Spoken Language Understanding with Regularized Representation Alignment
- Cross-lingual Text Classification with Heterogeneous Graph Neural Network