CLNov 14, 2019

Training a code-switching language model with monolingual data

arXiv:1911.06003v211 citations
AI Analysis

This addresses a data scarcity issue for researchers and practitioners working on multilingual NLP applications, though it appears incremental as it builds on existing RNN-based methods.

The paper tackles the problem of training code-switching language models without code-switching data by proposing an approach using monolingual data only, which improves performance and is comparable or better than using artificially generated data.

A lack of code-switching data complicates the training of code-switching (CS) language models. We propose an approach to train such CS language models on monolingual data only. By constraining and normalizing the output projection matrix in RNN-based language models, we bring embeddings of different languages closer to each other. Numerical and visualization results show that the proposed approaches remarkably improve the performance of CS language models trained on monolingual data. The proposed approaches are comparable or even better than training CS language models with artificially generated CS data. We additionally use unsupervised bilingual word translation to analyze whether semantically equivalent words in different languages are mapped together.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes