Pre-trained language models have been recently shown to benefit task-oriented dialogue (TOD) systems. Despite their success, existing methods often formulate this task as a cascaded generation problem which can lead to error accumulation across different sub-tasks and greater data annotation overhead. In this study, we present PPTOD, a unified plug-and-play model for task-oriented dialogue. In addition, we introduce a new dialogue multi-task pre-training strategy that allows the model to learn the primary TOD task completion skills from heterogeneous dialog corpora. We extensively test our model on three benchmark TOD tasks, including end-to-end dialogue modelling, dialogue state tracking, and intent classification. Experimental results show that PPTOD achieves new state of the art on all evaluated tasks in both high-resource and low-resource scenarios. Furthermore, comparisons against previous SOTA methods show that the responses generated by PPTOD are more factually correct and semantically coherent as judged by human annotators.

Contextual embedding-based language models trained on large data sets, such as BERT and RoBERTa, provide strong performance across a wide range of tasks and are ubiquitous in modern NLP. It has been observed that fine-tuning these models on tasks involving data from domains different from that on which they were pretrained can lead to suboptimal performance. Recent work has explored approaches to adapt pretrained language models to new domains by incorporating additional pretraining on domain-specific corpora and task data. We propose an alternative approach for transferring pretrained language models to new domains by adapting their tokenizers. We show that domain-specific subword sequences can be determined efficiently directly from divergences in the conditional token distributions of the base and domain-specific corpora. In datasets from four disparate domains, we find adaptive tokenization on a pretrained RoBERTa model provides greater than 85% of the performance benefits of domain specific pretraining. Our approach produces smaller models and less training and inference time than other approaches using tokenizer augmentation. Although using adaptive tokenization incurs a 6% increase in model parameters (due to the introduction of 10k new domain-specific tokens), our approach, using 64 CPUs, is >72x faster than further pretraining the language model on domain-specific corpora on 8 TPUs.

Behavior of deep neural networks can be inconsistent between different versions. Regressions during model update are a common cause of concern that often over-weigh the benefits in accuracy or efficiency gain. This work focuses on quantifying, reducing and analyzing regression errors in the NLP model updates. Using negative flip rate as regression measure, we show that regression has a prevalent presence across tasks in the GLUE benchmark. We formulate the regression-free model updates into a constrained optimization problem, and further reduce it into a relaxed form which can be approximately optimized through knowledge distillation training method. We empirically analyze how model ensemble reduces regression. Finally, we conduct CheckList behavioral testing to understand the distribution of regressions across linguistic phenomena, and the efficacy of ensemble and distillation methods.

Summarizing data samples by quantitative measures has a long history, with descriptive statistics being a case in point. However, as natural language processing methods flourish, there are still insufficient characteristic metrics to describe a collection of texts in terms of the words, sentences, or paragraphs they comprise. In this work, we propose metrics of diversity, density, and homogeneity that quantitatively measure the dispersion, sparsity, and uniformity of a text collection. We conduct a series of simulations to verify that each metric holds desired properties and resonates with human intuitions. Experiments on real-world datasets demonstrate that the proposed characteristic metrics are highly correlated with text classification performance of a renowned model, BERT, which could inspire future applications.

Pre-trained language models that learn contextualized word representations from a large un-annotated corpus have become a standard component for many state-of-the-art NLP systems. Despite their successful applications in various downstream NLP tasks, the extent of contextual impact on the word representation has not been explored. In this paper, we present a detailed analysis of contextual impact in Transformer- and BiLSTM-based masked language models. We follow two different approaches to evaluate the impact of context: a masking based approach that is architecture agnostic, and a gradient based approach that requires back-propagation through networks. The findings suggest significant differences on the contextual impact between the two model architectures. Through further breakdown of analysis by syntactic categories, we find the contextual impact in Transformer-based MLM aligns well with linguistic intuition. We further explore the Transformer attention pruning based on our findings in contextual analysis.

Hierarchical neural networks are often used to model inherent structures within dialogues. For goal-oriented dialogues, these models miss a mechanism adhering to the goals and neglect the distinct conversational patterns between two interlocutors. In this work, we propose Goal-Embedded Dual Hierarchical Attentional Encoder-Decoder (G-DuHA) able to center around goals and capture interlocutor-level disparity while modeling goal-oriented dialogues. Experiments on dialogue generation, response generation, and human evaluations demonstrate that the proposed model successfully generates higher-quality, more diverse and goal-centric dialogues. Moreover, we apply data augmentation via goal-oriented dialogue generation for task-oriented dialog systems with better performance achieved.