Open AkihikoWatanabe opened 1 year ago
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 Q2, 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 Q2 against other metrics using this dataset and two others, where it consistently shows higher correlation with human judgements.
(knowledge-grounded; 知識に基づいた)対話に対するFactual ConsistencyをReference-freeで評価できるQGQA手法。機械翻訳やAbstractive Summarizationの分野で研究が進んできたが、対話では
といった外部知識に対するconsistencyが適切ではない要素が多く存在し、よりチャレンジングなタスクとなっている。 また、そもそも対話タスクはopen-endedなタスクなため、Reference-basedな手法は現実的ではなく、Reference-freeな手法が必要と主張。
手法の概要としては以下。ユーザの発話からQuestion Generation (QG)を実施し、Question-Answer Candidate Pairを作成する。そして、生成したQuestionをベースとなる知識から回答させ(QA)、その回答結果とAnswer Candidateを比較することでFactual Consistencyを測定する。
https://aclanthology.org/2021.emnlp-main.619/