Code Switching
Code-switching, the alternation of languages within a single utterance, is a linguistic phenomenon increasingly studied using computational methods to understand its complexities and improve multilingual natural language processing. Current research focuses on developing robust models, often leveraging contrastive learning and mixture-of-experts architectures, to handle code-switched text and speech in various tasks like machine translation, speech recognition, and language identification. These advancements are crucial for creating more inclusive and effective technologies that cater to the diverse linguistic practices of multilingual communities, particularly in low-resource language settings.
Papers
Improving Code-switching Language Modeling with Artificially Generated Texts using Cycle-consistent Adversarial Networks
Chia-Yu Li, Ngoc Thang Vu
ASCEND: A Spontaneous Chinese-English Dataset for Code-switching in Multi-turn Conversation
Holy Lovenia, Samuel Cahyawijaya, Genta Indra Winata, Peng Xu, Xu Yan, Zihan Liu, Rita Frieske, Tiezheng Yu, Wenliang Dai, Elham J. Barezi, Qifeng Chen, Xiaojuan Ma, Bertram E. Shi, Pascale Fung