Retrieval Enhanced Data Augmentation for Question Answering on Privacy Policies

Retrieval Enhanced Data Augmentation for Question Answering on Privacy Policies
复制标题

用于隐私政策问答的检索增强数据增强

DOI:
10.48550/arxiv.2204.08952
复制
发表时间:
2022
期刊:
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Law
影响因子:
--
通讯作者:
Kai
Kai
中科院分区:
--
文献类型:
--
作者:
Md. Rizwan Parvez;Jianfeng Chi;Wasi Uddin Ahmad;Yuan Tian;Kai

文献摘要

参考文献

被引文献

相似文献

在隐私政策框架的问题回答(QA)的任务,以确定最相关的文本片段或一个列表的句子从给定的用户查询的政策文件之前的研究。现有的标记数据集是严重不平衡的(只有几个相关的部分),限制了QA性能在这个领域。在本文中,我们开发了一个数据增强框架的基础上集成检索模型,捕捉相关的文本片段从未标记的政策文件,并扩大训练集中的积极的例子。此外,为了提高增强数据的多样性和质量,我们利用了多个预训练的语言模型(LM),并将它们与降噪预言机级联。使用我们在PrivacyQA基准测试上的增强数据,我们大幅提升了现有的基线(10%F1),并实现了50%的新的最先进的F1得分。我们的消融研究为我们的方法的有效性提供了进一步的见解。
Prior studies in privacy policies frame the question answering (QA) task as identifying the most relevant text segment or a list of sentences from a policy document given a user query. Existing labeled datasets are heavily imbalanced (only a few relevant segments), limiting the QA performance in this domain. In this paper, we develop a data augmentation framework based on ensembling retriever models that captures the relevant text segments from unlabeled policy documents and expand the positive examples in the training set. In addition, to improve the diversity and quality of the augmented data, we leverage multiple pre-trained language models (LMs) and cascaded them with noise reduction oracles. Using our augmented data on the PrivacyQA benchmark, we elevate the existing baseline by a large margin (10% F1) and achieve a new state-of-the-art F1 score of 50%. Our ablation studies provide further insights into the effectiveness of our approach.
DOI: 10.18653/v1/2021.naacl-main.402
发表时间: 2021-04
期刊: --
影响因子: --
作者:
Md. Rizwan Parvez;Kai-Wei Chang
通讯作者: Md. Rizwan Parvez;Kai-Wei Chang