CoSDA-ML: Multi-Lingual Code-Switching Data Augmentation for Zero-Shot Cross-Lingual NLP
By Libo Qin, Minheng Ni, Yue Zhang, Wanxiang Che. 2020
173 citations recorded by the source corpus.
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.
Subfields: Artificial Intelligence, Natural Language Processing, Machine Translation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Jianfeng Gao
- Mohit Bansal
- Zhe Gan
- Percy Liang
- Noah A. Smith
- Luke S. Zettlemoyer
- Yejin Choi
- Graham Neubig
In Natural Language Processing
- Sebastian Gehrmann
- Jianfeng Gao
- Mohit Bansal
- Erik Cambria
- Graham Neubig
- Ming Zhou
- Yejin Choi
- Thomas Scialom
In Machine Translation
- Rui Wang
- Ananya B. Sai
- Jianfeng Gao
- Ondrej Dusek
- Kalpesh Krishna
- Sebastian Ruder
- Sebastian Gehrmann
- Nan Duan
Citations in the corpus
- Data Augmentation Approaches in Natural Language Processing: A Survey (2021, methodology): “[72] propose a multi-lingual code-switching method that replaces original words in the source language with words of other languages.”
- Syntax-Aware Data Augmentation for Neural Machine Translation (2020, methodology): “Some methods [7], [39], [44] use language model to create additional training data by calculating the similarities between words or generating the new representation of words.”
- M3P: Learning Universal Representations via Multitask Multilingual Multimodal Pre-training (2021, methodology): “Code-switched Training Code-switched training [27] [33] converts the original training corpus to code-switched corpus, which can help the model explicitly model the relationship among corresponding words in different languages.”
- Recent advances in deep learning based dialogue systems: a systematic survey (2021, background): “Actually, there are quite a few recent works in dialogue systems that train GANs with reinforcement learning framework (Zhu et al., 2018; Wu et al., 2019b; He et al., 2020a; Zhu et al., 2020; Qin et al., 2020).”
- NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation (2021, methodology): “This transformation generates multi-lingual code-switching data to fine-tune encoders of large language models (Qin et al., 2020; Tan and Joty, 2021; Wang et al., 2019b) by making use of bilingual dictionaries of MUSE (Lample et al., 2018).”