Context Of Dialogue Research Articles

Heterogeneous graph reasoning for knowledge-grounded medical dialogue system

Beyond the common difficulties faced in task-oriented dialogue system, medical dialogue has recently attracted increasing attention due to its huge application potential while posing more challenges in reasoning over medical domain knowledge and logic. Existing works resort to neural language models for dialogue embedding and neglect the explicit logical reasoning, leading to poor explainable and generalization ability. In this work, we propose an explainable Heterogeneous Graph Reasoning (HGR) model to unify the relational dialogue context understanding and entity-correlation reasoning into a heterogeneous graph structure. HGR encodes entity context according to the corresponding utterance and deduces next response after fusing the underlying medical knowledge with entity context by attentional graph propagation. To push forward the future research on expert-sensitive task-oriented dialogue system, we first release a large-scale Medical Dialogue Consultant benchmark (MDG-C) with 16 Gastrointestinal diseases for evaluating consultant capability and a Medical Dialogue Diagnosis benchmark (MDG-D) with 6 diseases for measuring diagnosis capability of models, respectively. Extensive experiments on both MDG-C and MDG-D benchmarks demonstrate the superiority of our HGR over state-of-the-art knowledge grounded approaches in general fields of medical dialogue system.

Learning to Respond with Your Favorite Stickers

Stickers with vivid and engaging expressions are becoming increasingly popular in online messaging apps, and some works are dedicated to automatically select sticker response by matching the stickers image with previous utterances. However, existing methods usually focus on measuring the matching degree between the dialog context and sticker image, which ignores the user preference of using stickers. Hence, in this article, we propose to recommend an appropriate sticker to user based on multi-turn dialog context and sticker using history of user. Two main challenges are confronted in this task. One is to model the sticker preference of user based on the previous sticker selection history. Another challenge is to jointly fuse the user preference and the matching between dialog context and candidate sticker into final prediction making. To tackle these challenges, we propose a Preference Enhanced Sticker Response Selector (PESRS) model. Specifically, PESRS first employs a convolutional-based sticker image encoder and a self-attention-based multi-turn dialog encoder to obtain the representation of stickers and utterances. Next, deep interaction network is proposed to conduct deep matching between the sticker and each utterance. Then, we model the user preference by using the recently selected stickers as input and use a key-value memory network to store the preference representation. PESRS then learns the short-term and long-term dependency between all interaction results by a fusion network and dynamically fuses the user preference representation into the final sticker selection prediction. Extensive experiments conducted on a large-scale real-world dialog dataset show that our model achieves the state-of-the-art performance for all commonly used metrics. Experiments also verify the effectiveness of each component of PESRS.

Enhancing the anti-domestic violence competencies of undergraduates: action research on a social work curriculum in Mainland China

ABSTRACT Based on a domestic violence intervention undergraduate course offered by a university in Mainland China, this study explores how dialogic education help enhance the competencies of students on domestic violence intervention through action research, with a key framework on gender sensitivity in a multicultural context. In the action process of dialogic education, teachers are the facilitators, constructing diverse dialogic contexts for students. This enables students to break the culture of silence about domestic violence in freewriting, understand intimate partner violence from a gender perspective, develop critical reflection through Chinese and international comparison view, and enhance cultural sensitivity through multicultural group interactions, all of which enable students to achieve professional competence in domestic violence intervention, based on their life experiences. Finally, this paper discusses the further application of dialogue-based anti-domestic violence education in the China context.

Predicting Emotion Reactions for Human–Computer Conversation: A Variational Approach

Although emotional conversation generation has attracted widespread attention in recent years, research works on the emotional interactional mechanism are still scarce, which makes it difficult for existing emotional dialogue system to automatically determine a suitable emotion type for conversation generation. Such a response emotion planning task is often difficult due to the “gap” problem: we need to predict the emotional probability distribution of the upcoming responses, which have not yet been generated. In this article, we propose a novel method, namely interactional emotion learning (IEL), which adopts an intuitive way to eliminate the “gap” problem: we design a variational learning network, called potential response learning, to learn the latent distribution of responses for given conversation in a semantic space. Then, we predict an appropriate emotion for response generation based on both the dialogue context and the learned latent distribution. Extensive experiments have been performed on three off-the-shelf conversation datasets, and the experimental results show that the proposed variational learning network significantly boosts the prediction ability of our approach, and our IEL method outperforms the state-of-the-art dialogue classification methods in the emotion planning task.

Medical Term and Status Generation From Chinese Clinical Dialogue With Multi-Granularity Transformer

This paper describes a generative model for extracting medical terms and their status from Chinese medical dialogues. Notably, the extracted semantic information is particularly important to downstream tasks like automatic medical scribe and automatic diagnosis systems. However, how to effectively leverage dialogue context to generate medical terms and their corresponding status accurately remains less explored. Existing generative approaches treat dialogue text as a single continuous text, ignoring conversational characteristics like colloquialism, redundancy and interactions. Between the doctor and the patient, a variety of colloquial medical information is frequently discussed. Each speaker (doctor and patient) plays a specific role in the interaction's goals. As a result, the importance of role information and interactions between utterances cannot be overstated. Furthermore, existing generative approaches only use character-level tokens, disregarding word-level tokens, which are the shortest meaningful utterances in Chinese. In this paper, we propose a Multi-granularity Transformer (MGT) model to enhance the dialogue context understanding from multi-granularity features. We incorporate word-level information by adapting a Lattice-based encoder with our proposed relative position encoding method. We further propose a Role Access Controlled Attention (RaCa) mechanism for introducing utterance-level interaction information. Experimental results on two benchmark datasets illustrate our model's validity and effectiveness, achieving state-of-the-art performance on both datasets.

Label and Context Augmentation for Response Selection at DSTC8

This paper studies the dialogue response selection task. As state-of-the-arts are neural models requiring a large training set, data augmentation has been considered as a means to overcome the sparsity of observational annotation, where only one observed response is annotated as gold. In this paper, we first consider label augmentation, of selecting, among unobserved utterances, that would “counterfactually” replace the labeled response, for the given context, and augmenting labels only if that is the case. The key advantage of this model is not incurring human annotation overhead, thus not increasing the training cost, i.e., for low-resource scenarios. In addition, we consider context augmentation scenarios where the given dialogue context is not sufficient for label augmentation. In this case, inspired by open-domain question answering, we “decontextualize” by retrieving missing contexts, such as related persona. We empirically show that our pipeline improves BERT-based models in two different response selection tasks without incurring annotation overheads.

Multi-Turn Dialogue Reading Comprehension With Pivot Turns and Knowledge

Multi-turn dialogue reading comprehension aims to teach machines to read dialogue contexts and solve tasks such as response selection and answering questions. The major challenges involve noisy history contexts and especial prerequisites of commonsense knowledge that is unseen in the given material. Existing works mainly focus on context and response matching approaches. This work thus makes the first attempt to tackle the above two challenges by extracting substantially important turns as pivot utterances and utilizing external knowledge to enhance the representation of context. We propose a pivot-oriented deep selection model (PoDS) on top of the Transformer-based language models for dialogue comprehension. In detail, our model first picks out the pivot utterances from the conversation history according to the semantic matching with the candidate response or question, if any. Besides, knowledge items related to the dialogue context are extracted from a knowledge graph as external knowledge. Then, the pivot utterances and the external knowledge are combined together with a well-designed mechanism for refining predictions. Experimental results on four dialogue comprehension benchmark tasks show that our proposed model achieves great improvements on baselines. A series of empirical comparisons are conducted to show how our selection strategies and the extra knowledge injection influence the results.

Conversational Semantic Role Labeling

Semantic role labeling (SRL) aims to extract the arguments for each predicate in an input sentence. Traditional SRL can fail to analyze dialogues because it only works on every single sentence, while ellipsis and anaphora frequently occur in dialogues. To address this problem, we propose the conversational SRL task, where an argument can be the dialogue participants, a phrase in the dialogue history or the current sentence. As the existing SRL datasets are in the sentence level, we manually annotate semantic roles for 3000 chit-chat dialogues (27198 sentences) to boost the research in this direction. Experiments show that while traditional SRL systems (even with the help of coreference resolution or rewriting) perform poorly for analyzing dialogues, modeling dialogue histories and participants greatly helps the performance, indicating that adapting SRL to conversations is very promising for universal dialogue understanding. Our initial study by applying CSRL to two mainstream conversational tasks, dialogue response generation and dialogue context rewriting, also confirms the usefulness of CSRL.

Knowing Where to Leverage: Context-Aware Graph Convolutional Network With an Adaptive Fusion Layer for Contextual Spoken Language Understanding

Spoken language understanding (SLU) systems aim to understand users’ utterance, which is a key component of task-oriented dialogue systems. In this paper, we focus on improving the contextual SLU. The contextual SLU systems mainly focus on how to effectively incorporate dialog context information (contextual information). The existing approaches all use the same contextual information to guide slot filling at all tokens, which may inject the irrelevant information and result in ambiguity. To tackle this problem, we propose a context-aware graph convolutional network (GCN) with an adaptive fusion layer for contextual SLU. The context-aware GCN is proposed to automatically aggregate the contextual information, which frees our model from the manually designed heuristic aggregation function. Meanwhile, an adaptive fusion layer is applied at each token to dynamically incorporate relevant contextual information, which achieves a fine-grained contextual information transfer to guide the token-level slot filling. Experiments on the Simulated Dialog Dataset show that our model achieves state-of-the-art performance and outperforms other previous methods by a large margin (+3.67% on Sim-R, +4.18% on Sim-M and +3.75% on Overall dataset). In addition, we explore and analyze the pre-trained model (i.e., BERT) in our framework. We show that incorporating BERT brings a large improvement in low-resource setting.

New Pages of S.L. Frank's Biography in the Context of the Dialogue between Russian and Western Philosophical Cultures

-

Interreligious Dialogue in Context

Abstract The article sketches the overall layout of the thematic issue of the ‘Journal of Religion and Transformation in Contemporary Societies (JRAT)’ on Interreligious Dialogue (IRD) in context. It argues that an analysis of Interreligious Dialogue-activities in their socio-cultural contexts helps to counterbalance the long-standing individualistic bias of IRD-research. First, it presents a systematic description of the present state of the art that distinguishes two strands of IRD-research. Second, it argues for a European comparison, based upon the most recent findings from the ‘SMRE – Swiss Metadatabase of Religious Affiliation in Europe’. The article closes with references to the structure of the present volume of JRaT to facilitate such a comparison.

Legitimising Religion in Public

Abstract The English context for interreligious dialogue is shaped by the presence of an established church which is inclusive, geographically spread, and engages with the state. This article will trace the ways in which the presence of an established church, and the particular model of church-state settlement, provide a context to legitimise particular types of interreligious activity. The social role of religion, the representative function of religion, and religion as an inclusive category, will be highlighted as key elements in the role of religion in English public life and in how interreligious organisations have developed. This observation is analytically useful as it assists an understanding of how and why interreligious dialogue and other activity has at various points become significant for the state’s governance of religious diversity, how success is understood and managed, and what non-engagement with interreligious activity might indicate.

Qur'ān as a Hidden Academy for Learning Dialogic Exchange

The Qur’ān hosts numerous dialogues independent of its readers’ moral and ideological positions – a key source for renewed learning and understanding. Conventional readings of the Qur’ān rarely notice the subtle underlying linguistic features. Similarly, rote readings can narrow one’s capacity for renewed learning. As a result, the accustomed reader misses the profound moral lessons. To address this problem, I have analysed how the Qur’ān can be imagined as an open academy for learning dialogic exchange. This article’s thematic approach is inspired by heuristic modes of learning to ascertain the Qur’ān’s dialogic medium as a model of ‘invitational rhetoric.’
 The theories in this article create an inter-disciplinary framework. I used the Arabic word qawl (statement/assertion) as a primary linguistical index of dialogic exchange. I quantitatively surveyed two morphological derivatives of this word across the Qur’ānic text: qul (say, denoting God’s voice) and qālū (they said, plural in the past tense, denoting voices of the Other vis-à-vis God). These derivatives signify voices speaking from different yet interactive contexts.
 This article argues the Qur’ān’s down-to-earth guidance translates the notion of dialogue from an elitist noun concept into an interactive dialogic context. The Qur’ān’s dialogic medium illustrates its artistic elevation of dialogic exchanges at two levels: justice and iḥsān (excellence). To this end, I use content analysis to analyse emergent themes according to the three principles of invitational rhetoric. The outcomes of this analysis ascertained how the Qur’ān records multiple voices of dialogic exchange, which it accords with its aesthetic features.

The Use of Dialogic Principles of Romanian Crowdfunding Platform Communication on Facebook

This study examines how the Romanian crowdfunding platform sprijina.ro is using social media, specifically Facebook, in order to create a dialogic context for consumers, as defined by the five dialogic principles issued by Kent and Taylor (1998). To determine if dialogic communica- tion was occurring between the organization and its followers, the content analysis method was applied to the Facebook page (posts, comments and reactions) of sprijina.ro crowdfunding platform over a 2 years span (2017– 2019). The study attempts to answer the question of whether Romanian crowdfunding platform sprijina.ro applies the principles of dialogic com- munication on their Facebook page. The results show that when applying the dialogic principles, the online audience become more engaged. Unfor- tunately, the majority of posts analyzed ask for money donation without taking into consideration public interest or trying to engage public in the online discussion. This type of communication reflects the lack of strategy in creating and maintaining a dialogic relationship with the online audi- ence through Facebook posts.

A step forward in understandingpas: the post-verbal negator in Old Occitan from the perspective of communication and rhetorical strategizing

AbstractIn recent years, within the cognitive linguistics approach there has been a trend of scholarly research committed to exploring the motivation for language change. The way in which people use language in communication, together with principles of human categorization, are the locus where language change and innovations are to be found. Interactional contexts in particular, seen as playing a crucial role in bringing about syntactic change (Traugott 2010b), have figured prominently in recent contributions on diachronic micro-changes, bringing to the fore the role played by dialogue as both the manifestation of the participants’ own voices and the realization of the constant negotiation that characterizes human communication. Against this background, this contribution focuses on a particular use of the Occitan post-verbal negatorpasin negative rhetorical questions, which was very productive in fifteenth-century collections of religious theatrical texts. It is claimed that these dialogic contexts allowed a polyphonic use ofpas, crucially restricted to this post-verbal negator, which is key to identifying the reasons behind the eventual establishment ofpasas the generalized sentential negator in the modern language.

Philosophy (and argumentation) for children: some reflection for primary school

Through the exercise of inquiry and thinking together, a number of studies promote philosophy at school within the dialogical context. The Philosophy for Children curriculum and other Philosophy with Children experiences all around the world spearheaded on promoting dialogical contexts in school with the aim of foster a high-order thinking. The P4C became the standard reference for revision of curricula and teaching practices, on the one hand, through enhancing thinking and reasoning skills, and, on the other, by promote a democratic space and the exercise of citizenship. With a complex - thinking model and with a community of inquiry framework, Lipman (2003) upholds the idea that thinking does not consist of just reasoning or logic itself, but it is performed by different dimensions of thought, intra- and inter-personal. The thinking space that a philosophical discussion could in fact open supports the exercise of reasoning in communities of inquiry that promote cognitive, social and civic competence. Acknowledging the crucial role of the rational and social dimensions of thinking, scholarly research that investigates “argumentation” in children and in school settings is growing. Consistently with studies in Argumentation and Education , in this paper we will refer to “argumentation” as a fundamental way of reasoning and a social practice which finds the ideal context in which it can develop in the discussion (Mirza, Perret-Clermont, 2009). Philosophical and psycho-pedagogical studies recognize argumentation as an essential actor in the construction of thought and in the structuring of new knowledge, in content understanding and in the creation of contextual connections, in the dynamic of relationship, in the engagement of various metacognitive processes (Rapanta et al., 2013). Rational and social aspects of argumentative competence could be fostered in a philosophical context at primary school age. In consideration of the need to promote social and civic competences and based on the definition of discussion as the ideal context to promote argumentation and to practice democracy, it becomes important to raise teacher’s awareness of the key importance of argumentation. Recognizing some relevant perspectives in Argumentation and Education , this paper attempts to highlight several theoretical and methodological questions that are relevant to Philosophy for/with Children and to teacher and facilitator training.

Robust dialog state tracker with contextual-feature augmentation

Dialog state tracking (DST), which estimates dialog states given a dialog context, is a core component in task-oriented dialog systems. Existing data-driven methods usually extract features automatically through deep learning. However, most of these models have limitations. First, compared with hand-crafted delexicalization features, such features in deep learning approaches are not universal. However, they are important for tracking unseen slot values. Second, such models do not work well in situations where noisy labels are ubiquitous in datasets. To address these challenges, we propose a robust dialog state tracker with contextual-feature augmentation. Contextual-feature augmentation is used to extract generalized features; hence, it is capable of solving the unseen slot value tracking problem. We apply a simple but effective deep learning paradigm to train our DST model with noisy labels. The experimental results show that our model achieves state-of-the-art scores in terms of joint accuracy on the MultiWOZ 2.0 dataset. In addition, we show its performance in tracking unseen slot values by simulating unseen domain dialog state tracking.

The Role of the English Article System in Developing Dialogical Context: A View from Russian Science

A common issue arising in international scientific communication concerns the use of zero-, indefinite- and definite articles in English. While existing pedagogical approaches are successfully used to teach near-native competency in the use of English articles, the final stage of native-equivalent competency continues to evade even advanced ESL / EAP students, especially those whose first language (e.g. Russian) does not possess articles. Therefore, the present work is aimed at developing a conceptual approach to article use based on a consideration of the development of context in scientific communication. A literature review of some existing theoretical approaches shows a clear development from semantic and transformational-grammatical attitudes towards more pragmatic explicatory strategies based on dialogic communication. A qualitative content analysis of article errors appearing in the text of research papers written by Russian scientists and corrected by a native English editor revealed the presence of nine major article-related error types, of which the use of zero article with singular NPs instead of the definite article (SxO√D) was the most common. NPs in sentences containing article errors were also evaluated according to Bickerton’s NP semantic function typology, Hawkins’ location theory and Liu and Gleason’s major types. The hypothesis that Russian scientists, being highly educated, would generally tend to overuse the, a and an in an overcompensation for the tendency of less-well-educated Russians to drop articles when using English was not confirmed by the results of the content analysis. The analysis of article-use error types appearing in different sections of an IMRaD research paper showed that errors related to the overuse and underuse of the definite article the are particularly characteristic of the Introduction section. However, the largest number of article-related errors were found in the Discussion section, where the SxO√D error significantly outweighed other error types as compared with other sections. While existing conceptual approaches to explicating the function of the English article system have limited utility, a comprehensive system has yet to be developed. The authors recommend a dialogic approach for teaching the English article system to non-native speakers in the context of scientific communication.

The Role of the English Article System in Developing Dialogical Context: A View from Russian Science

A common issue arising in international scientific communication concerns the use of zero-, indefinite- and definite articles in English. While existing pedagogical approaches are successfully used to teach near-native competency in the use of English articles, the final stage of native-equivalent competency continues to evade even advanced ESL / EAP students, especially those whose first language (e.g. Russian) does not possess articles. Therefore, the present work is aimed at developing a conceptual approach to article use based on a consideration of the development of context in scientific communication. A literature review of some existing theoretical approaches shows a clear development from semantic and transformational-grammatical attitudes towards more pragmatic explicatory strategies based on dialogic communication. A qualitative content analysis of article errors appearing in the text of research papers written by Russian scientists and corrected by a native English editor revealed the presence of nine major article-related error types, of which the use of zero article with singular NPs instead of the definite article (SxØ√D) was the most common. NPs in sentences containing article errors were also evaluated according to Bickerton’s NP semantic function typology, Hawkins’ location theory and Liu and Gleason’s major types. The hypothesis that Russian scientists, being highly educated, would generally tend to overuse the, a and an in an overcompensation for the tendency of less-well-educated Russians to drop articles when using English was not confirmed by the results of the content analysis. The analysis of article-use error types appearing in different sections of an IMRaD research paper showed that errors related to the overuse and underuse of the definite article the are particularly characteristic of the Introduction section. However, the largest number of article-related errors were found in the Discussion section, where the SxØ√D error significantly outweighed other error types as compared with other sections. While existing conceptual approaches to explicating the function of the English article system have limited utility, a comprehensive system has yet to be developed. The authors recommend a dialogic approach for teaching the English article system to non-native speakers in the context of scientific communication.

Context-Controlled Topic-Aware Neural Response Generation for Open-Domain Dialog Systems

Incorporating topic information can help response generation models to produce informative responses for chat-bots. Previous work only considers the individual semantic of each topic, ignoring its specific dialog context, which may result in inaccurate topic representation and hurt response coherence. Besides, as an important feature of multi-turn conversation, dynamic topic transitions have not been well-studied. We propose a Context-Controlled Topic-Aware neural response generation model, i.e., CCTA, which makes dialog context interact with the process of topic representing and transiting to achieve balanced improvements on response informativeness and contextual coherence. CCTA focuses on capturing the semantical relations within topics as well as their corresponding contextual information in conversation, to produce context-dependent topic representations at the word-level and turn-level. Besides, CCTA introduces a context-controlled topic transition strategy, utilizing contextual topics to yield relevant transition words. Extensive experimental results on two benchmark multi-turn conversation datasets validate the superiority of our proposal on generating coherent and informative responses against the state-of-the-art baselines. We also find that topic transition modeling can work as an auxiliary learning task to boost the response generation.