Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders

@inproceedings{Zhao2017LearningDD,
  title={Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders},
  author={Tiancheng Zhao and Ran Zhao and Maxine Esk{\'e}nazi},
  booktitle={ACL},
  year={2017}
}
While recent neural encoder-decoder models have shown great promise in modeling open-domain conversations, they often generate dull and generic responses. [...] Key Method Our model uses latent variables to learn a distribution over potential conversational intents and generates diverse responses using only greedy decoders. We have further developed a novel variant that is integrated with linguistic prior knowledge for better performance.Expand
Hierarchy Response Learning for Neural Conversation Generation
TLDR
A hierarchical response generation (HRG) framework is proposed to capture the conversation intention in a natural and coherent way and can generate the responses with more appropriate content and expression. Expand
BERT for Open-Domain Conversation Modeling
TLDR
It is demonstrated that simply using fixed pre-trained BERT as part of the model without further finetuning is powerful enough for generating better responses in terms of fluency, grammar, and semantic coherency. Expand
DialogWAE: Multimodal Response Generation with Conditional Wasserstein Auto-Encoder
TLDR
DialogWAE is proposed, a conditional Wasserstein autoencoder specially designed for dialogue modeling that models the distribution of data by training a GAN within the latent variable space and develops a Gaussian mixture prior network to enrich the latent space. Expand
Improving Variational Encoder-Decoders in Dialogue Generation
TLDR
A separate VED model is developed that learns to autoencode discrete texts into continuous embeddings and generalize latent representations by reconstructing the encoded embedding through transforming Gaussian noise through multi-layer perceptrons. Expand
Jointly Optimizing Diversity and Relevance in Neural Response Generation
TLDR
A SpaceFusion model to jointly optimize diversity and relevance that essentially fuses the latent space of a sequence-to-sequence model and that of an autoencoder model by leveraging novel regularization terms is proposed. Expand
mu-Forcing: Training Variational Recurrent Autoencoders for Text Generation
TLDR
The proposed method directly injects extra constraints on the posteriors of latent variables into the learning process of VRAE, which can flexibly and stably control the trade-off between the KL term and the reconstruction term, making the model learn dense and meaningful latent representations. Expand
A Discrete CVAE for Response Generation on Short-Text Conversation
TLDR
A discrete latent variable with an explicit semantic meaning to improve the CVAE on short-text conversation and proposes a two-stage sampling approach to enable efficient diverse variable selection from a large latent space assumed in the short- text conversation task. Expand
Better Conversations by Modeling, Filtering, and Optimizing for Coherence and Diversity
TLDR
A measure of coherence is introduced as the GloVe embedding similarity between the dialogue context and the generated response to improve coherence and diversity in encoder-decoder models for open-domain conversational agents. Expand
MEMD: A Diversity-Promoting Learning Framework for Short-Text Conversation
TLDR
An easy-to-extend learning framework named MEMD (Multi-Encoder to Multi-Decoder), in which an auxiliary encoder and an auxiliary decoder are introduced to provide necessary training guidance without resorting to extra data or complicating network’s inner structure is proposed. Expand
A Hierarchical Latent Structure for Variational Conversation Modeling
TLDR
A novel model named Variational Hierarchical Conversation RNNs (VHCR), involving two key ideas of using a hierarchical structure of latent variables, and exploiting an utterance drop regularization is proposed, which successfully utilizes latent variables and outperforms state-of-the-art models for conversation generation. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 47 REFERENCES
A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues
TLDR
A neural network-based generative architecture, with latent stochastic variables that span a variable number of time steps, that improves upon recently proposed models and that the latent variables facilitate the generation of long outputs and maintain the context. Expand
Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models
TLDR
The recently proposed hierarchical recurrent encoder-decoder neural network is extended to the dialogue domain, and it is demonstrated that this model is competitive with state-of-the-art neural language models and back-off n-gram models. Expand
A Diversity-Promoting Objective Function for Neural Conversation Models
TLDR
This work proposes using Maximum Mutual Information (MMI) as the objective function in neural models, and demonstrates that the proposed MMI models produce more diverse, interesting, and appropriate responses, yielding substantive gains in BLEU scores on two conversational datasets and in human evaluations. Expand
Generating Sentences from a Continuous Space
TLDR
This work introduces and study an RNN-based variational autoencoder generative model that incorporates distributed latent representations of entire sentences that allows it to explicitly model holistic properties of sentences such as style, topic, and high-level syntactic features. Expand
Deep Reinforcement Learning for Dialogue Generation
TLDR
This work simulates dialogues between two virtual agents, using policy gradient methods to reward sequences that display three useful conversational properties: informativity, non-repetitive turns, coherence, and ease of answering. Expand
A Neural Conversational Model
TLDR
A simple approach to conversational modeling which uses the recently proposed sequence to sequence framework, and is able to extract knowledge from both a domain specific dataset, and from a large, noisy, and general domain dataset of movie subtitles. Expand
Towards End-to-End Learning for Dialog State Tracking and Management using Deep Reinforcement Learning
This paper presents an end-to-end framework for task-oriented dialog systems using a variant of Deep Recurrent Q-Networks (DRQN). The model is able to interface with a relational database and jointlyExpand
Topic Augmented Neural Response Generation with a Joint Attention Mechanism
TLDR
A topic augmented joint attention based Seq2Seq (TAJA-Seq 2Seq) model that simulates how people behave in conversation and can generate well-focused and informative responses with the help of topic information is proposed. Expand
Learning Structured Output Representation using Deep Conditional Generative Models
TLDR
A deep conditional generative model for structured output prediction using Gaussian latent variables is developed, trained efficiently in the framework of stochastic gradient variational Bayes, and allows for fast prediction using Stochastic feed-forward inference. Expand
Sequence to Sequence Learning with Neural Networks
TLDR
This paper presents a general end-to-end approach to sequence learning that makes minimal assumptions on the sequence structure, and finds that reversing the order of the words in all source sentences improved the LSTM's performance markedly, because doing so introduced many short term dependencies between the source and the target sentence which made the optimization problem easier. Expand
...
1
2
3
4
5
...