Nearest Neighbour Few-Shot Learning for Cross-lingual Classification

Abstract

Even though large pre-trained multilingual models (e.g. mBERT, XLM-R) have led to significant performance gains on a wide range of cross-lingual NLP tasks, success on many downstream tasks still relies on the availability of sufficient annotated data. Traditional fine-tuning of pre-trained models using only a few target samples can cause over-fitting. This can be quite limiting as most languages in the world are under-resourced. In this work, we investigate cross-lingual adaptation using a simple nearest neighbor few-shot (<15 samples) inference technique for classification tasks. We experiment using a total of 16 distinct languages across two NLP tasks- XNLI and PAWS-X. Our approach consistently improves traditional fine-tuning using only a handful of labeled samples in target locales. We also demonstrate its generalization capability across tasks.

References in corpus (6)

• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• Cross-Lingual Ability of Multilingual BERT: An Empirical Study
• Self-training Improves Pre-training for Natural Language Understanding
• From Zero to Hero: On the Limitations of Zero-Shot Cross-Lingual Transfer with Multilingual Transformers
• $k$-Nearest Neighbor Augmented Neural Networks for Text Classification
• Paraphrase Augmented Task-Oriented Dialog Generation