Artificial Code-Switching for Cross-Lingual Embedding Alignment

Using task-specific pre-training and leveraging cross-lingual transfer are two of the most popular ways to handle code-switched data. In this paper, we aim to compare the effects of both for the task of sentiment analysis. We work with two Dravidian Code-Switched languages - Tamil-Engish and Malayalam-English and four different BERT based models. We compare the effects of task-specific pre-training and cross-lingual transfer and find that task-specific pre-training results in superior zero-shot and supervised performance when compared to performance achieved by leveraging cross-lingual transfe

Research goal: Does incorporating artificially code-switched data during pre-training improve the alignment of multilingual embedding spaces for cross-lingual semantic textual similarity tasks compared to standard parallel data augmentation?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10.