Meta learning to classify intent and slot labels with noisy few shot examples
This work is significant for developers of spoken language understanding systems, as it provides a method to improve model performance and robustness when training data is noisy and scarce, which is a common real-world problem.
The paper introduces a new benchmarking task, few-shot robust spoken language understanding (SLU), to address the challenge of training SLU models with noisy and limited data. They propose a novel meta-learning model based on prototypical networks that consistently outperforms fine-tuning and MAML baselines in intent classification accuracy and slot labeling F1, while also showing smaller performance variation under noise.
Recently deep learning has dominated many machine learning areas, including spoken language understanding (SLU). However, deep learning models are notorious for being data-hungry, and the heavily optimized models are usually sensitive to the quality of the training examples provided and the consistency between training and inference conditions. To improve the performance of SLU models on tasks with noisy and low training resources, we propose a new SLU benchmarking task: few-shot robust SLU, where SLU comprises two core problems, intent classification (IC) and slot labeling (SL). We establish the task by defining few-shot splits on three public IC/SL datasets, ATIS, SNIPS, and TOP, and adding two types of natural noises (adaptation example missing/replacing and modality mismatch) to the splits. We further propose a novel noise-robust few-shot SLU model based on prototypical networks. We show the model consistently outperforms the conventional fine-tuning baseline and another popular meta-learning method, Model-Agnostic Meta-Learning (MAML), in terms of achieving better IC accuracy and SL F1, and yielding smaller performance variation when noises are present.