Leveraging representations from pre-trained transformer-based encoders achieves state-of-the-art performance on numerous NLP tasks. Larger encoders can improve accuracy for spoken language understanding (SLU) but are challenging to use given the inference latency constraints of online systems (especially on CPU machines).We evaluate using a larger 170M parameter BERT encoder that shares representations across languages, domains and tasks for SLU compared to using smaller 17M parameter BERT encoders with language-, domain- and task-decoupled finetuning.Running inference with a larger shared encoder on GPU is latency neutral and reduces infrastructure cost compared to running inference for decoupled smaller encoders on CPU machines. The larger shared encoder reduces semantic error rates by 4.62% for test sets representing user requests to voice-controlled devices and 5.79% on the tail of the test sets on average across four languages.

Spoken Language Understanding (SLU) models in industry applications are usually trained offline on historic data, but have to perform well on incoming user requests after deployment. Since the application data is not available at training time, this is formally similar to the domain generalization problem, where domains correspond to different temporal segments of the data, and the goal is to build a model that performs well on unseen domains, e.g., upcoming data. In this paper, we explore different strategies for achieving good temporal generalization, including instance weighting, temporal fine-tuning, learning temporal features and building a temporally-invariant model. Our results on data of large-scale SLU systems show that temporal information can be leveraged to improve temporal generalization for SLU models.

In this study, we investigate robustness against covariate drift in spoken language understanding (SLU). Covariate drift can occur in SLUwhen there is a drift between training and testing regarding what users request or how they request it. To study this we propose a method that exploits natural variations in data to create a covariate drift in SLU datasets. Experiments show that a state-of-the-art BERT-based model suffers performance loss under this drift. To mitigate the performance loss, we investigate distributionally robust optimization (DRO) for finetuning BERT-based models. We discuss some recent DRO methods, propose two new variants and empirically show that DRO improves robustness under drift.

In productionized machine learning systems, online model performance is known to deteriorate over time when there is a distributional drift between offline training and online application data. As a remedy, models are typically retrained at fixed time intervals, implying high computational and manual costs. This work aims at decreasing such costs in productionized, large-scale Spoken Language Understanding systems. In particular, we develop a need-based re-training strategy guided by an efficient drift detector and discuss the arising challenges including system complexity, overlapping model releases, observation limitation and the absence of annotated resources at runtime. We present empirical results on historical data and confirm the utility of our design decisions via an online A/B experiment.

In this paper, we explore class-incremental learning for intent classification (IC) in a setting with limited old data available. IC is the task of mapping user utterances to their corresponding intents. Even though class-incremental learning without storing the old data yields high potential of reducing human and computational resources in industry NLP model releases, to the best of our knowledge, it hasn’t been studied for NLP classification tasks in the literature before. In this work, we compare several contemporary class-incremental learning methods, i.e., BERT warm start, L2, Elastic Weight Consolidation, RecAdam and Knowledge Distillation within two realistic class-incremental learning scenarios: one where only the previous model is assumed to be available, but no data corresponding to old classes, and one in which limited unlabeled data for old classes is assumed to be available. Our results indicate that among the investigated continual learning methods, Knowledge Distillation worked best for our class-incremental learning tasks, and adding limited unlabeled data helps the model in both adaptability and stability.

This paper provides the first experimental study on the impact of using domain-specific representations on a BERT-based multi-task spoken language understanding (SLU) model for multi-domain applications. Our results on a real-world dataset covering three languages indicate that by using domain-specific representations learned adversarially, model performance can be improved across all of the three SLU subtasks domain classification, intent classification and slot filling. Gains are particularly large for domains with limited training data.

This paper addresses the question as to what degree a BERT-based multilingual Spoken Language Understanding (SLU) model can transfer knowledge across languages. Through experiments we will show that, although it works substantially well even on distant language groups, there is still a gap to the ideal multilingual performance. In addition, we propose a novel BERT-based adversarial model architecture to learn language-shared and language-specific representations for multilingual SLU. Our experimental results prove that the proposed model is capable of narrowing the gap to the ideal multilingual performance.

This work explores cross-lingual transfer learning (TL) for named entity recognition, focusing on bootstrapping Japanese from English. A deep neural network model is adopted and the best combination of weights to transfer is extensively investigated. Moreover, a novel approach is presented that overcomes linguistic differences between this language pair by romanizing a portion of the Japanese input. Experiments are conducted on external datasets, as well as internal large-scale real-world ones. Gains with TL are achieved for all evaluated cases. Finally, the influence on TL of the target dataset size and of the target tagset distribution is further investigated.

A typical cross-lingual transfer learning approach boosting model performance on a language is to pre-train the model on all available supervised data from another language. However, in large-scale systems this leads to high training times and computational requirements. In addition, characteristic differences between the source and target languages raise a natural question of whether source data selection can improve the knowledge transfer. In this paper, we address this question and propose a simple but effective language model based source-language data selection method for cross-lingual transfer learning in large-scale spoken language understanding. The experimental results show that with data selection i) source data and hence training speed is reduced significantly and ii) model performance is improved.

This paper investigates the use of Machine Translation (MT) to bootstrap a Natural Language Understanding (NLU) system for a new language for the use case of a large-scale voice-controlled device. The goal is to decrease the cost and time needed to get an annotated corpus for the new language, while still having a large enough coverage of user requests. Different methods of filtering MT data in order to keep utterances that improve NLU performance and language-specific post-processing methods are investigated. These methods are tested in a large-scale NLU task with translating around 10 millions training utterances from English to German. The results show a large improvement for using MT data over a grammar-based and over an in-house data collection baseline, while reducing the manual effort greatly. Both filtering and post-processing approaches improve results further.