Retrieve and Refine: Improved Sequence Generation Models For Dialogue
arXiv:1808.04776
Abstract
Sequence generation models for dialogue are known to have several problems: they tend to produce short, generic sentences that are uninformative and unengaging. Retrieval models on the other hand can surface interesting responses, but are restricted to the given retrieval set leading to erroneous replies that cannot be tuned to the specific context. In this work we develop a model that combines the two approaches to avoid both their deficiencies: first retrieve a response and then refine it -- the final sequence generator treating the retrieval as additional context. We show on the recent CONVAI2 challenge task our approach produces responses superior to both standard retrieval and generation models in human evaluations.
References in corpus (12)
- Sequence to Sequence Learning with Neural Networks
- How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
- Personalizing Dialogue Agents: I have a dog, do you have pets too?
- An Information Retrieval Approach to Short Text Conversation
- Pre-Translation for Neural Machine Translation
- Generative Deep Neural Networks for Dialogue: A Short Review
- Search Engine Guided Non-Parametric Neural Machine Translation
- Generating Sentences by Editing Prototypes
- Controllable Abstractive Summarization
- Iterative Refinement for Machine Translation
- An Exploration of Neural Sequence-to-Sequence Architectures for Automatic Post-Editing
- QuickEdit: Editing Text & Translations by Crossing Words Out
Cited by in corpus (11)
- Neural Text Generation with Unlikelihood Training
- Explaining and Improving Model Behavior with k Nearest Neighbor Representations
- Nearest Neighbor Machine Translation
- Getting To Know You: User Attribute Extraction from Dialogues
- Skeleton-to-Response: Dialogue Generation Guided by Retrieval Memory
- Shaping the Narrative Arc: An Information-Theoretic Approach to Collaborative Dialogue
- I like fish, especially dolphins: Addressing Contradictions in Dialogue Modeling
- Data-to-Text Generation with Style Imitation
- Label-Agnostic Sequence Labeling by Copying Nearest Neighbors
- Response Selection for Multi-Party Conversations with Dynamic Topic Tracking
- Predict and Use Latent Patterns for Short-Text Conversation