Flowchart-grounded troubleshooting dialogue (FTD) systems, which follow the instructions of a flowchart to diagnose users' problems in specific domains (e.g., vehicle, laptop), have been gaining research interest in recent years. However, collecting sufficient dialogues that are naturally grounded on flowcharts is costly, thus FTD systems are impeded by scarce training data. To mitigate the data sparsity issue, we propose a plan-based synthetic data generation (PlanSDG) approach that generates diverse synthetic dialog data at scale by transforming concise flowchart into dialogues. Specifically, its generative model employs a variational-base framework with a hierarchical planning strategy that includes global and local latent planning variables. Experiments on the FloDial dataset show that synthetic dialogue produced by PlanSDG improves the performance of downstream tasks, including flowchart path retrieval and response generation, in particular on the Out-of-Flowchart settings. In addition, further analysis demonstrate the quality of synthetic data generated by PlanSDG in paths that are covered by current sample dialogues and paths that are not covered.

Flowchart-grounded troubleshooting dialogue (FTD) systems, which follow the instructions of a flowchart to diagnose users’ problems in specific domains (e.g., vehicle, laptop), have been gaining research interest in recent years. However, collecting sufficient dialogues that are naturally grounded on flowcharts is costly, thus FTD systems are impeded by scarce training data. To mitigate the data sparsity issue, we propose a plan-based data augmentation ( PlanDA ) approach that generates diverse synthetic dialog data at scale by transforming concise flowchart into dialogues. Specifically, its generative model employs a variational-base framework with a hierarchical planning strategy that includes global and local latent planning variables. Experiments on the FloDial dataset show that synthetic dialogue produced by PlanDA improves the performance of down-stream tasks, including flowchart path retrieval and response generation, in particular on the Out-of-Flowchart settings. In addition, further analysis demonstrate the quality of synthetic data generated by PlanDA in paths that are covered by current sample dialogues and paths that are not covered.

We consider the influence of two types of contextual information, background information available to users and users' goals, on users' views and preferences regarding textual explanations generated for the outcomes predicted by Decision Trees (DTs). To investigate the influence of background information, we generate contrastive explanations that address potential con-flicts between aspects of DT predictions and plausible expectations licensed by background information. We define four types of conflicts, operationalize their identification, and specify explanatory schemas that address them. To investigate the influence of users' goals, we employ an interactive setting where given a goal and an initial explanation for a predicted outcome, users select follow-up questions, and assess the explanations that answer these questions. Here, we offer algorithms to generate explanations that address six types of follow-up questions. The main result from both user studies is that explanations which have a contrastive aspect about a predicted class are generally preferred by users. In addition, the results from the first study indicate that these explanations are deemed especially valuable when users expectations differ from predicted outcomes; and the results from the second study indicate that contrastive explanations which describe how to change a predicted outcome are particularly well regarded in terms of helping users achieve this goal, and they are also popular in terms of helping users achieve other goals.

Lifelong Learning (LL) black-box models are dynamic in that they keep learning from new tasks and constantly update their parameters. Owing to the need to utilize information from previously seen tasks, and capture commonalities in potentially diverse data, it is hard for automatic explanation methods to explain the outcomes of these models. In addition, existing explanation methods, e.g., LIME, which are computationally expensive when explaining a static black-box model, are even more inefficient in the LL setting. In this paper, we propose a novel Lifelong Explanation (LLE) approach that continuously trains a student explainer under the supervision of a teacher – an arbitrary explanation algorithm – on different tasks undertaken in LL. We also leverage the Experience Replay (ER) mechanism to prevent catastrophic forgetting in the student explainer. Our experiments comparing LLE to three baselines on text classification tasks show that LLE can enhance the stability of the explanations for all seen tasks and maintain the same level of faithfulness to the black-box model as the teacher, while being up to 10ˆ2 times faster at test time. Our ablation study shows that the ER mechanism in our LLE approach enhances the learning capabilities of the student explainer. Our code is available at https://github.com/situsnow/LLE.

We offer an approach to explain Decision Tree (DT) predictions by addressing potential conflicts between aspects of these predictions and plausible expectations licensed by background information. We define four types of conflicts, operationalize their identification, and specify explanatory schemas that address them. Our human evaluation focused on the effect of explanations on users’ understanding of a DT’s reasoning and their willingness to act on its predictions. The results show that (1) explanations that address potential conflicts are considered at least as good as baseline explanations that just follow a DT path; and (2) the conflict-based explanations are deemed especially valuable when users’ expectations disagree with the DT’s predictions.

Machine translation (MT) is an important task in natural language processing (NLP), as it automates the translation process and reduces the reliance on human translators. With the resurgence of neural networks, the translation quality surpasses that of the translations obtained using statistical techniques for most language-pairs. Up until a few years ago, almost all of the neural translation models translated sentences independently , without incorporating the wider document-context and inter-dependencies among the sentences. The aim of this survey article is to highlight the major works that have been undertaken in the space of document-level machine translation after the neural revolution, so researchers can recognize the current state and future directions of this field. We provide an organization of the literature based on novelties in modelling and architectures as well as training and decoding strategies. In addition, we cover evaluation strategies that have been introduced to account for the improvements in document MT, including automatic metrics and discourse-targeted test sets. We conclude by presenting possible avenues for future exploration in this research field.

Xuelin Situ, Ingrid Zukerman, Cecile Paris, Sameen Maruf, Gholamreza Haffari. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

We report the results of the first edition of the WMT shared task on chat translation. The task consisted of translating bilingual conversational text, in particular customer support chats for the English-German language pair (English agent, German customer). This task varies from the other translation shared tasks, i.e. news and biomedical, mainly due to the fact that the conversations are bilingual, less planned, more informal, and often ungrammatical. Furthermore, such conversations are usually characterized by shorter and simpler sentences and contain more pronouns. We received 14 submissions from 6 participating teams, all of them covering both directions, i.e. En->De for agent utterances and De->En for customer messages. We used automatic metrics (BLEU and TER) for evaluating the translations of both agent and customer messages and human document-level direct assessments (DDA) to evaluate the agent translations.

The advent of context-aware NMT has resulted in promising improvements in the overall translation quality and specifically in the translation of discourse phenomena such as pronouns. Previous works have mainly focused on the use of past sentences as context with a focus on anaphora translation. In this work, we investigate the effect of future sentences as context by comparing the performance of a contextual NMT model trained with the future context to the one trained with the past context. Our experiments and evaluation, using generic and pronoun-focused automatic metrics, show that the use of future context not only achieves significant improvements over the context-agnostic Transformer, but also demonstrates comparable and in some cases improved performance over its counterpart trained on past context. We also perform an evaluation on a targeted cataphora test suite and report significant gains over the context-agnostic Transformer in terms of BLEU.

Despite the progress made in sentence-level NMT, current systems still fall short at achieving fluent, good quality translation for a full document. Recent works in context-aware NMT consider only a few previous sentences as context and may not scale to entire documents. To this end, we propose a novel and scalable top-down approach to hierarchical attention for context-aware NMT which uses sparse attention to selectively focus on relevant sentences in the document context and then attends to key words in those sentences. We also propose single-level attention approaches based on sentence or word-level information in the context. The document-level context representation, produced from these attention modules, is integrated into the encoder or decoder of the Transformer model depending on whether we use monolingual or bilingual context. Our experiments and evaluation on English-German datasets in different document MT settings show that our selective attention approach not only significantly outperforms context-agnostic baselines but also surpasses context-aware baselines in most cases.

We describe the work of Monash University for the shared task of Rotowire document translation organised by the 3rd Workshop on Neural Generation and Translation (WNGT 2019). We submitted systems for both directions of the English-German language pair. Our main focus is on employing an established document-level neural machine translation model for this task. We achieve a BLEU score of 39.83 (41.46 BLEU per WNGT evaluation) for En-De and 45.06 (47.39 BLEU per WNGT evaluation) for De-En translation directions on the Rotowire test set. All experiments conducted in the process are also described.

Recent works in neural machine translation have begun to explore document translation. However, translating online multi-speaker conversations is still an open problem. In this work, we propose the task of translating Bilingual Multi-Speaker Conversations, and explore neural architectures which exploit both source and target-side conversation histories for this task. To initiate an evaluation for this task, we introduce datasets extracted from Europarl v7 and OpenSubtitles2016. Our experiments on four language-pairs confirm the significance of leveraging conversation history, both in terms of BLEU and manual evaluation.

We present a document-level neural machine translation model which takes both source and target document context into account using memory networks. We model the problem as a structured prediction problem with interdependencies among the observed and hidden variables, i.e., the source sentences and their unobserved target translations in the document. The resulting structured prediction problem is tackled with a neural translation model equipped with two memory components, one each for the source and target side, to capture the documental interdependencies. We train the model end-to-end, and propose an iterative decoding algorithm based on block coordinate descent. Experimental results of English translations from French, German, and Estonian documents show that our model is effective in exploiting both source and target document context, and statistically significantly outperforms the previous work in terms of BLEU and METEOR.

Feature selection (FS) is employed to make text classification (TC) more effective. Well-known FS metrics like information gain (IG) and odds ratio (OR) rank terms without considering term interactions. Building classifiers with FS algorithms considering term interactions can yield better performance. But their computational complexity is a concern. This has resulted in two-stage algorithms such as information gain-principal component analysis (IG–PCA). Random forests-based feature selection (RFFS), proposed by Breiman, has demonstrated outstanding performance while capturing gene–gene relations in bioinformatics, but its usefulness for TC is less explored. RFFS has fewer control parameters and is found to be resistant to overfitting and thus generalizes well to new data. It does not require use of a test dataset to report accuracy and does not use conventional cross-validation. This paper investigates the working of RFFS for TC and compares its performance against IG, OR and IG–PCA. We carry out experiments on four widely used text data sets using naive Bayes’ and support vector machines as classifiers. RFFS achieves macro-F 1 values higher than other FS algorithms in 73 % of the experimental instances. We also analyze the performance of RFFS for TC in terms of its parameters and class skews of the data sets and yield interesting results.

Designing a good feature selection (FS) algorithm is of utmost importance especially for text classification (TC), wherein a large number of features representing terms or words pose serious challenges to the effectiveness and efficiency of classifiers. FS algorithms are divided into two broad categories, namely, feature ranking (FR) and feature subset selection (FSS) algorithms. Unlike FSS, FR algorithms select those features that are individually highly relevant for the class or category without taking the feature interactions into account. This makes FR algorithms simple and computationally more efficient than FSS and thus, mostly a preferred choice for TC. Bi-normal separation (BNS) (Forman, 2003) and information gain (IG) (Yang and Pedersen, 1997) are well-known FR metrics. However, FR algorithms output a set of highly relevant features or terms which can possibly be redundant and can thus, deteriorate a classifier׳s performance. This paper suggests taking the interactions of words into account in order to eliminate redundant terms. Stand-alone FSS algorithms can be computationally expensive for the high-dimensional text data. We therefore suggest a two-stage FS algorithm, which employs an FR metric such as BNS or IG in the first stage and an FSS algorithm such as the Markov blanket filter (MBF) (Koller and Sahami, 1996) in the second stage. Most of the two-stage algorithms proposed in the literature for TC combine feature ranking and feature transformation such as principal component analysis (PCA) algorithms. To estimate the statistical significance of our two-stage algorithm, we carry out experiments on 10 different splits of training and test sets of each of the three (Reuters-21578, TREC, OHSUMED) data sets with naive Bayes׳ and support vector machines. Our results based on a paired two-sided t-test show that the macro F1 performance of BNS+MBF is statistically significant than that of stand-alone BNS in 69% of the total experimental trials. The macro F1 values of IG get enhanced in 72% of the trials when MBF is used in the second stage. We also compare our two-stage algorithm against two recently proposed FS algorithms, namely, distinguishing feature selector (DFS) (Uysal and Gunal, 2012) and a two stage algorithm consisting of IG and PCA algorithms (Uguz, 2011). BNS+MBF is found to be significantly better than DFS and IG+PCA in 74 and 78% of the trials respectively. IG+MBF outperforms DFS and IG+PCA in 93 and 80% of the experimental trials respectively. Similar results are observed for BNS+MBF and IG+MBF when the performances are evaluated in terms of balanced error rate.