Two are Better than One: An Ensemble of Retrieval- and Generation-Based Dialog Systems

Abstract

Open-domain human-computer conversation has attracted much attention in the field of NLP. Contrary to rule- or template-based domain-specific dialog systems, open-domain conversation usually requires data-driven approaches, which can be roughly divided into two categories: retrieval-based and generation-based systems. Retrieval systems search a user-issued utterance (called a query) in a large database, and return a reply that best matches the query. Generative approaches, typically based on recurrent neural networks (RNNs), can synthesize new replies, but they suffer from the problem of generating short, meaningless utterances. In this paper, we propose a novel ensemble of retrieval-based and generation-based dialog systems in the open domain. In our approach, the retrieved candidate, in addition to the original query, is fed to an RNN-based reply generator, so that the neural model is aware of more information. The generated reply is then fed back as a new candidate for post-reranking. Experimental results show that such ensemble outperforms each single part of it by a large margin.

References in corpus (10)

• ADADELTA: An Adaptive Learning Rate Method
• A Neural Conversational Model
• How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
• Neural Responding Machine for Short-Text Conversation
• A Hierarchical Neural Autoencoder for Paragraphs and Documents
• An Information Retrieval Approach to Short Text Conversation
• A Network-based End-to-End Trainable Task-oriented Dialogue System
• Continuously Learning Neural Dialogue Management
• Stochastic Language Generation in Dialogue using Recurrent Neural Networks with Convolutional Sentence Reranking
• StalemateBreaker: A Proactive Content-Introducing Approach to Automatic Human-Computer Conversation

Cited by in corpus (23)

• A Survey on Dialogue Systems: Recent Advances and New Frontiers
• Emotional Chatting Machine: Emotional Conversation Generation with Internal and External Memory
• Meta-evaluation of Conversational Search Evaluation Metrics
• Deep Learning Based Chatbot Models
• Generating Sentences by Editing Prototypes
• Neural Language Generation: Formulation, Methods, and Evaluation
• Towards Retrieval-based Conversational Recommendation
• Augmenting Transformers with KNN-Based Composite Memory for Dialogue
• Say What I Want: Towards the Dark Side of Neural Dialogue Models
• Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory
• RubyStar: A Non-Task-Oriented Mixture Model Dialog System
• Keep CALM and Explore: Language Models for Action Generation in Text-based Games
• An Enhanced Knowledge Injection Model for Commonsense Generation
• Text Generation with Exemplar-based Adaptive Decoding
• Extract and Edit: An Alternative to Back-Translation for Unsupervised Neural Machine Translation
• Retrieval-based Goal-Oriented Dialogue Generation
• Weakly-Supervised Neural Response Selection from an Ensemble of Task-Specialised Dialogue Agents
• Iterative Refinement in the Continuous Space for Non-Autoregressive Neural Machine Translation
• Data-Efficient Methods for Dialogue Systems
• Predict and Use Latent Patterns for Short-Text Conversation
• Data-to-text Generation by Splicing Together Nearest Neighbors
• Augmenting Machine Learning with Information Retrieval to Recommend Real Cloned Code Methods for Code Completion
• QuaSE: Accurate Text Style Transfer under Quantifiable Guidance