arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2106.04016cs.CL
收藏

Disfl-QA:一个理解问答中不流利现象的基准数据集

Disfl-QA: A Benchmark Dataset for Understanding Disfluencies in Question Answering

Google(谷歌);The University of Texas at Austin(德克萨斯大学奥斯汀分校);Georgia Institute of Technology(佐治亚理工学院)

Aditya Gupta, Jiacheng Xu, Shyam Upadhyay, Diyi Yang, Manaal Faruqui

摘要:

不流利是NLP中研究不足的主题,尽管它在人类对话中无处不在。这主要是由于缺乏包含不流利内容的数据集。在本文中,我们提出了一个新的挑战问答数据集,Disfl-QA,SQuAD的衍生物,人类在以前流利的问题中引入上下文不流利。Disfl-QA包含各种具有挑战性的不流利,需要比以前的数据集更全面地理解文本。实验表明,现有的国家的最先进的问答模型的性能显着下降时,测试在zero-shot setting.We显示数据增强方法部分恢复的性能损失,也证明了使用黄金数据进行微调的有效性。我们认为,我们需要大规模的不流利数据集,以便NLP模型对它们具有鲁棒性。该数据集可在www.example.com上公开获得。

英文摘要:

Disfluencies is an under-studied topic in NLP, even though it is ubiquitous in human conversation. This is largely due to the lack of datasets containing disfluencies. In this paper, we present a new challenge question answering dataset, Disfl-QA, a derivative of SQuAD, where humans introduce contextual disfluencies in previously fluent questions. Disfl-QA contains a variety of challenging disfluencies that require a more comprehensive understanding of the text than what was necessary in prior datasets. Experiments show that the performance of existing state-of-the-art question answering models degrades significantly when tested on Disfl-QA in a zero-shot setting.We show data augmentation methods partially recover the loss in performance and also demonstrate the efficacy of using gold data for fine-tuning. We argue that we need large-scale disfluency datasets in order for NLP models to be robust to them. The dataset is publicly available at: https://github.com/google-research-datasets/disfl-qa.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑