Multilingual Code-Switching for Zero-Shot Cross-Lingual Intent Prediction and Slot Filling

TL;DR

This paper introduces a novel multilingual code-switching data augmentation method to improve zero-shot cross-lingual intent prediction and slot filling, demonstrating significant accuracy gains across multiple languages and real-world crisis data.

Contribution

The study proposes a new data augmentation technique using multilingual code-switching to enhance language neutrality in transformers for zero-shot NLU tasks.

Findings

Achieved +4.2% accuracy in intent prediction across 8 languages.

Improved F1 score by +1.8% for slot filling.

Validated effectiveness on a new Haitian Creole-English dataset during a disaster.

Abstract

Predicting user intent and detecting the corresponding slots from text are two key problems in Natural Language Understanding (NLU). In the context of zero-shot learning, this task is typically approached by either using representations from pre-trained multilingual transformers such as mBERT, or by machine translating the source data into the known target language and then fine-tuning. Our work focuses on a particular scenario where the target language is unknown during training. To this goal, we propose a novel method to augment the monolingual source data using multilingual code-switching via random translations to enhance a transformer's language neutrality when fine-tuning it for a downstream task. This method also helps discover novel insights on how code-switching with different language families around the world impact the performance on the target language. Experiments on the benchmark dataset of MultiATIS++ yielded an average improvement of +4.2% in accuracy for intent task and +1.8% in F1 for slot task using our method over the state-of-the-art across 8 different languages. Furthermore, we present an application of our method for crisis informatics using a new human-annotated tweet dataset of slot filling in English and Haitian Creole, collected during Haiti earthquake disaster.