$Q^2$: Evaluating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answering

Abstract

Neural knowledge-grounded generative models for dialogue often produce content that is factually inconsistent with the knowledge they rely on, making them unreliable and limiting their applicability. Inspired by recent work on evaluating factual consistency in abstractive summarization, we propose an automatic evaluation metric for factual consistency in knowledge-grounded dialogue using automatic question generation and question answering. Our metric, denoted $Q^2$, compares answer spans using natural language inference (NLI), instead of token-based matching as done in previous work. To foster proper evaluation, we curate a novel dataset of dialogue system outputs for the Wizard-of-Wikipedia dataset, manually annotated for factual consistency. We perform a thorough meta-evaluation of $Q^2$ against other metrics using this dataset and two others, where it consistently shows higher correlation with human judgements.

arXiv commentAccepted to EMNLP 2021

References in corpus (9)

• BERTScore: Evaluating Text Generation with BERT
• Wizard of Wikipedia: Knowledge-Powered Conversational agents
• How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation
• Towards a Human-like Open-Domain Chatbot
• Evaluation of Text Generation: A Survey
• Faithful to the Original: Fact Aware Neural Abstractive Summarization
• Assessing The Factual Accuracy of Generated Text
• QuestEval: Summarization Asks for Fact-based Evaluation
• Improving Factual Consistency of Abstractive Summarization via Question Answering