Wizard of Wikipedia: Knowledge-Powered Conversational agents

Abstract

In open-domain dialogue intelligent agents should exhibit the use of knowledge, however there are few convincing demonstrations of this to date. The most popular sequence to sequence models typically "generate and hope" generic utterances that can be memorized in the weights of the model when mapping from input utterance(s) to output, rather than employing recalled knowledge as context. Use of knowledge has so far proved difficult, in part because of the lack of a supervised learning benchmark task which exhibits knowledgeable open dialogue with clear grounding. To that end we collect and release a large dataset with conversations directly grounded with knowledge retrieved from Wikipedia. We then design architectures capable of retrieving knowledge, reading and conditioning on it, and finally generating natural responses. Our best performing dialogue models are able to conduct knowledgeable discussions on open-domain topics as evaluated by automatic metrics and human evaluations, while our new benchmark allows for measuring further improvements in this important research direction.

References in corpus (12)

• A Neural Conversational Model
• Universal Sentence Encoder
• SQuAD: 100,000+ Questions for Machine Comprehension of Text
• A Simple, Fast Diverse Decoding Algorithm for Neural Generation
• Efficient Natural Language Response Suggestion for Smart Reply
• Personalizing Dialogue Agents: I have a dog, do you have pets too?
• A Network-based End-to-End Trainable Task-oriented Dialogue System
• StarSpace: Embed All The Things!
• A Neural Network Approach to Context-Sensitive Generation of Conversational Responses
• Generative Deep Neural Networks for Dialogue: A Short Review
• Training Millions of Personalized Dialogue Agents
• Extending Neural Generative Conversational Model using External Knowledge Sources

Cited by in corpus (87)

• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• A Survey of Knowledge-Enhanced Text Generation
• Autoregressive Entity Retrieval
• Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring
• QA Dataset Explosion: A Taxonomy of NLP Resources for Question Answering and Reading Comprehension
• Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue
• Recipes for Safety in Open-domain Chatbots
• Woodpecker: Hallucination Correction for Multimodal Large Language Models
• ACUTE-EVAL: Improved Dialogue Evaluation with Optimized Questions and Multi-turn Comparisons
• CorpusBrain: Pre-train a Generative Retrieval Model for Knowledge-Intensive Language Tasks
• Zero-Resource Knowledge-Grounded Dialogue Generation
• Meta-evaluation of Conversational Search Evaluation Metrics
• Challenges in Building Intelligent Open-domain Dialog Systems
• Proactive Human-Machine Conversation with Explicit Conversation Goals
• Towards Continual Knowledge Learning of Language Models
• A Survey of Natural Language Generation Techniques with a Focus on Dialogue Systems - Past, Present and Future Directions
• Knowledge-Grounded Dialogue Flow Management for Social Robots and Conversational Agents
• NaturalConv: A Chinese Dialogue Dataset Towards Multi-turn Topic-driven Conversation
• WikiChat: Stopping the Hallucination of Large Language Model Chatbots by Few-Shot Grounding on Wikipedia
• Efficient Retrieval Augmented Generation from Unstructured Knowledge for Task-Oriented Dialog
• Learning to Detect Relevant Contexts and Knowledge for Response Selection in Retrieval-based Dialogue Systems
• PLATO-2: Towards Building an Open-Domain Chatbot via Curriculum Learning
• Exploring Prompt-based Few-shot Learning for Grounded Dialog Generation
• Learning to Select Knowledge for Response Generation in Dialog Systems
• Bridging Text and Video: A Universal Multimodal Transformer for Video-Audio Scene-Aware Dialog
• EVA2.0: Investigating Open-Domain Chinese Dialogue Systems with Large-Scale Pre-Training
• Unstructured Text Enhanced Open-domain Dialogue System: A Systematic Survey
• Gunrock 2.0: A User Adaptive Social Conversational System
• An Empirical Investigation of Pre-Trained Transformer Language Models for Open-Domain Dialogue Generation
• Don't Say That! Making Inconsistent Dialogue Unlikely with Unlikelihood Training
• Augmenting Transformers with KNN-Based Composite Memory for Dialogue
• Incremental Transformer with Deliberation Decoder for Document Grounded Conversations
• PLATO: Pre-trained Dialogue Generation Model with Discrete Latent Variable
• Search-Engine-augmented Dialogue Response Generation with Cheaply Supervised Query Production
• Learning a Simple and Effective Model for Multi-turn Response Generation with Auxiliary Tasks
• Generate, Delete and Rewrite: A Three-Stage Framework for Improving Persona Consistency of Dialogue Generation
• Policy-Driven Neural Response Generation for Knowledge-Grounded Dialogue Systems
• Knowledge Aware Conversation Generation with Explainable Reasoning over Augmented Graphs
• KECRS: Towards Knowledge-Enriched Conversational Recommendation System
• A Survey of Document Grounded Dialogue Systems (DGDS)
• Personal Entity, Concept, and Named Entity Linking in Conversations
• Evaluating Empathetic Chatbots in Customer Service Settings
• ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning
• I love your chain mail! Making knights smile in a fantasy game world: Open-domain goal-oriented dialogue agents
• Thinking Globally, Acting Locally: Distantly Supervised Global-to-Local Knowledge Selection for Background Based Conversation
• RefNet: A Reference-aware Network for Background Based Conversation
• MoEL: Mixture of Empathetic Listeners
• Knowledge Injection into Dialogue Generation via Language Models
• Conditional Text Generation for Harmonious Human-Machine Interaction
• SideControl: Controlled Open-domain Dialogue Generation via Additive Side Networks
• Action-Based Conversations Dataset: A Corpus for Building More In-Depth Task-Oriented Dialogue Systems
• How To Evaluate Your Dialogue System: Probe Tasks as an Alternative for Token-level Evaluation Metrics
• Knowledge-Grounded Dialogue Generation with Pre-trained Language Models
• Conversations with Search Engines: SERP-based Conversational Response Generation
• A Document-grounded Matching Network for Response Selection in Retrieval-based Chatbots
• A Large-Scale Chinese Short-Text Conversation Dataset
• Proto: A Neural Cocktail for Generating Appealing Conversations
• Assessing Political Prudence of Open-domain Chatbots
• Baleen: Robust Multi-Hop Reasoning at Scale via Condensed Retrieval
• An Evaluation Protocol for Generative Conversational Systems
• Towards Socially Intelligent Agents with Mental State Transition and Human Utility
• Stay Hungry, Stay Focused: Generating Informative and Specific Questions in Information-Seeking Conversations
• Context Retrieval via Normalized Contextual Latent Interaction for Conversational Agent
• TempEL: Linking Dynamically Evolving and Newly Emerging Entities
• The StatCan Dialogue Dataset: Retrieving Data Tables through Conversations with Genuine Intents
• CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLP
• NewsDialogues: Towards Proactive News Grounded Conversation
• A Simple and Efficient Multi-Task Learning Approach for Conditioned Dialogue Generation
• Pretrained Language Models for Dialogue Generation with Multiple Input Sources
• Local Knowledge Powered Conversational Agents
• $Q^2$: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answering
• Filtering Noisy Dialogue Corpora by Connectivity and Content Relatedness
• WikiContradiction: Detecting Self-Contradiction Articles on Wikipedia
• Proposal Towards a Personalized Knowledge-powered Self-play Based Ensemble Dialog System
• Towards a Universal NLG for Dialogue Systems and Simulators with Future Bridging
• Multi-Task Learning of Generation and Classification for Emotion-Aware Dialogue Response Generation
• Know Deeper: Knowledge-Conversation Cyclic Utilization Mechanism for Open-domain Dialogue Generation
• Knowledge-driven Answer Generation for Conversational Search
• Positioning yourself in the maze of Neural Text Generation: A Task-Agnostic Survey
• POSSCORE: A Simple Yet Effective Evaluation of Conversational Search with Part of Speech Labelling
• Conversational Semantic Role Labeling
• Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation
• Would you Like to Talk about Sports Now? Towards Contextual Topic Suggestion for Open-Domain Conversational Agents
• A Corpus of Controlled Opinionated and Knowledgeable Movie Discussions for Training Neural Conversation Models
• Probing Commonsense Explanation in Dialogue Response Generation
• Viola: A Topic Agnostic Generate-and-Rank Dialogue System
• Extract, Denoise and Enforce: Evaluating and Improving Concept Preservation for Text-to-Text Generation