Hierarchical Variational Memory Network for Dialogue Generation

Chen, Hongshen; Zhang, Ren; Tang, Jiliang; Zhao, Yihong; Yin, Dawei

doi:10.1145/3178876.3186077

Cited by 78 publications

(52 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the purpose of exposition, we have chosen to avoid open questions that do not constrain user response for now. Even interpreting user responses to such questions is considered a challenging task [5]. learning, optimizing the reward of shorter turns and successful recommendations based on the FM's estimation of user preferred items and attributes, and the dialogue history.…”

Section: Introductionmentioning

confidence: 99%

Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Lei

Miao

et al. 2020

Proceedings of the 13th International Conference on Web Search and Data Mining

186

387

View full text Add to dashboard Cite

Recommender systems are embracing conversational technologies to obtain user preferences dynamically, and to overcome inherent limitations of their static models. A successful Conversational Recommender System (CRS) requires proper handling of interactions between conversation and recommendation. We argue that three fundamental problems need to be solved: 1) what questions to ask regarding item attributes, 2) when to recommend items, and 3) how to adapt to the users' online feedback. To the best of our knowledge, there lacks a unified framework that addresses these problems.In this work, we fill this missing interaction framework gap by proposing a new CRS framework named Estimation-Action-Reflection, or EAR, which consists of three stages to better converse with users. (1) Estimation, which builds predictive models to estimate user preference on both items and item attributes; (2) Action, which learns a dialogue policy to determine whether to ask attributes or recommend items, based on Estimation stage and conversation history; and (3) Reflection, which updates the recommender model when a user rejects the recommendations made by the Action stage. We present two conversation scenarios on binary and enumerated questions, and conduct extensive experiments on two datasets from Yelp and LastFM, for each scenario, respectively. Our experiments demonstrate significant improvements over the state-of-the-art method CRM [32], corresponding to fewer conversation turns and a higher level of recommendation hits. CCS CONCEPTS• Information systems → Users and interactive retrieval; Recommender systems; Personalization; • Human-centered computing → Interactive systems and tools.

show abstract

Section: Introductionmentioning

confidence: 99%

Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Lei

Miao

et al. 2020

Proceedings of the 13th International Conference on Web Search and Data Mining

186

387

View full text Add to dashboard Cite

show abstract

“…Neural dialogue generation aims at generating natural-sounding replies automatically to exchange information, e.g., knowledge [7,42,58]. As a core component of both task-oriented and non-taskoriented dialogue systems, neural dialogue generation has received a lot attention in recent years [1,3,36,58].…”

Section: Neural Dialogue Generationmentioning

confidence: 99%

“…Serban et al [38] utilize a latent variable at the sub-sequence level in a hierarchical setting. Chen et al [7] add a hierarchical structure and a variational memory module into a neural encoder-decoder network. However, all these latent vectors and latent memories are unexplainable, which makes it challenging to verify the effectiveness of dialogue state tracking.…”

Section: Dialogue State Trackingmentioning

confidence: 99%

See 1 more Smart Citation

Explicit State Tracking with Semi-Supervisionfor Neural Dialogue Generation

Jin

Lei

Zhang

et al. 2018

Proceedings of the 27th ACM International Conference on Information and Knowledge Management

Self Cite

View full text Add to dashboard Cite

The task of dialogue generation aims to automatically provide responses given previous utterances. Tracking dialogue states is an important ingredient in dialogue generation for estimating users' intention. However, the expensive nature of state labeling and the weak interpretability make the dialogue state tracking a challenging problem for both task-oriented and non-task-oriented dialogue generation: For generating responses in task-oriented dialogues, state tracking is usually learned from manually annotated corpora, where the human annotation is expensive for training; for generating responses in non-task-oriented dialogues, most of existing work neglects the explicit state tracking due to the unlimited number of dialogue states.In this paper, we propose the semi-supervised explicit dialogue state tracker (SEDST) for neural dialogue generation. To this end, our approach has two core ingredients: CopyFlowNet and posterior regularization. Specifically, we propose an encoder-decoder architecture, named CopyFlowNet, to represent an explicit dialogue state with a probabilistic distribution over the vocabulary space. To optimize the training procedure, we apply a posterior regularization strategy to integrate indirect supervision. Extensive experiments conducted on both task-oriented and non-task-oriented dialogue corpora demonstrate the effectiveness of our proposed model. Moreover, we find that our proposed semi-supervised dialogue state tracker achieves a comparable performance as state-of-the-art supervised learning baselines in state tracking procedure.

show abstract

“…The recurrent neural networks (RNNs) based sequence-to-sequence (Seq2Seq) architecture enables us to adopt natural text generation in a scalable and end-to-end manner. Many applications such as dialogue systems [2,11,13,25,26,32,35], document summarization [21] and poem generation [37,38] have been developed in the paradigm of Seq2Seq architecture.…”

Section: Introductionmentioning

confidence: 99%

Review-Driven Answer Generation for Product-Related Questions in E-Commerce

Chen

et al. 2019

Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining

View full text Add to dashboard Cite

The users often have many product-related questions before they make a purchase decision in E-commerce. However, it is often timeconsuming to examine each user review to identify the desired information. In this paper, we propose a novel review-driven framework for answer generation for product-related questions in Ecommerce, named RAGE. We develope RAGE on the basis of the multi-layer convolutional architecture to facilitate speed-up of answer generation with the parallel computation. For each question, RAGE first extracts the relevant review snippets from the reviews of the corresponding product. Then, we devise a mechanism to identify the relevant information from the noise-prone review snippets and incorporate this information to guide the answer generation. The experiments on two real-world E-Commerce datasets show that the proposed RAGE significantly outperforms the existing alternatives in producing more accurate and informative answers in natural language. Moreover, RAGE takes much less time for both model training and answer generation than the existing RNN based generation models.

show abstract

Hierarchical Variational Memory Network for Dialogue Generation

Cited by 78 publications

References 25 publications

Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Estimation-Action-Reflection: Towards Deep Interaction Between Conversational and Recommender Systems

Explicit State Tracking with Semi-Supervisionfor Neural Dialogue Generation

Review-Driven Answer Generation for Product-Related Questions in E-Commerce

Contact Info

Product

Resources

About