Semi-Supervised Few-Shot Intent Classification and Slot Filling

Abstract

Intent classification (IC) and slot filling (SF) are two fundamental tasks in modern Natural Language Understanding (NLU) systems. Collecting and annotating large amounts of data to train deep learning models for such systems is not scalable. This problem can be addressed by learning from few examples using fast supervised meta-learning techniques such as prototypical networks. In this work, we systematically investigate how contrastive learning and unsupervised data augmentation methods can benefit these existing supervised meta-learning pipelines for jointly modelled IC/SF tasks. Through extensive experiments across standard IC/SF benchmarks (SNIPS and ATIS), we show that our proposed semi-supervised approaches outperform standard supervised meta-learning methods: contrastive losses in conjunction with prototypical networks consistently outperform the existing state-of-the-art for both IC and SF tasks, while data augmentation strategies primarily improve few-shot IC by a significant margin.

TK-KNN: A Balanced Distance-Based Pseudo Labeling Approach for Semi-Supervised Intent Classification

“…Indeed, research most closely realted to the present work is the Slot-List model by Basu et al [2021], which focuses on the meta-learning aspect of semi-supervised learning rather than using unlabeled data. In a similar vein the GAN-BERT [Croce et al, 2020] model shows that using an adversarial learning regime can be devised to ensure that the extracted BERT features are similar amongst the unlabeled and the labeled data sets and substantially boost classification performance.…”