Open-domain clarification question generation without question examples
Open-domain clarification question generation without question examples
复制标题
没有问题示例的开放域澄清问题生成
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Noah D. Goodman
中科院分区：
文献类型：
--
作者：
Julia White;Gabriel Poesia;Robert D. Hawkins;Dorsa Sadigh;Noah D. Goodman
自然语言处理的首要目标是使机器能够与人类无缝沟通。然而，自然语言可能是模糊的或不清楚的。在不确定的情况下，人类会参与一个被称为修复的互动过程：提出问题并寻求澄清，直到他们的不确定性得到解决。我们提出了一个框架，建立一个视觉接地的提问模型，能够产生极性（是-否）澄清问题，以解决对话中的误解。我们的模型使用一个预期的信息增益目标，从现成的图像字幕，而不需要任何监督问答数据来获得翔实的问题。我们展示了我们的模型的能力，提出问题，提高沟通的成功，在一个目标导向的20个问题的游戏与合成和人类的回答。
An overarching goal of natural language processing is to enable machines to communicate seamlessly with humans. However, natural language can be ambiguous or unclear. In cases of uncertainty, humans engage in an interactive process known as repair: asking questions and seeking clarification until their uncertainty is resolved. We propose a framework for building a visually grounded question-asking model capable of producing polar (yes-no) clarification questions to resolve misunderstandings in dialogue. Our model uses an expected information gain objective to derive informative questions from an off-the-shelf image captioner without requiring any supervised question-answer data. We demonstrate our model’s ability to pose questions that improve communicative success in a goal-oriented 20 questions game with synthetic and human answerers.