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 2109.07009cs.CL
收藏

这个问题会不会很尴尬?基于答案模型提取的高效问答问题过滤

Will this Question be Answered? Question Filtering via Answer Model Distillation for Efficient Question Answering

Amazon Alexa AI(亚马逊 Alexa 人工智能)

Siddhant Garg, Alessandro Moschitti

摘要:

在本文中,我们提出了一种新的方法来提高效率的问答(QA)系统过滤掉的问题,将不会回答他们。这是基于一个有趣的新发现:最先进的QA系统的答案置信度分数可以通过仅使用输入问题文本的模型很好地近似。这使得能够抢先过滤由于其答案置信度分数低于系统阈值而未被系统回答的问题。具体来说,我们通过提取基于transformer的回答模型来学习基于transformer的问题模型。我们在三个流行的QA数据集和一个工业QA基准测试上的实验表明,我们的问题模型能够很好地近似目标QA系统的精度/召回曲线。这些问题模型在用作过滤器时,可以有效地权衡QA系统的较低计算成本以获得较低的召回率,例如,计算量减少约60%,而召回率仅损失约3-4%。

英文摘要:

In this paper we propose a novel approach towards improving the efficiency of Question Answering (QA) systems by filtering out questions that will not be answered by them. This is based on an interesting new finding: the answer confidence scores of state-of-the-art QA systems can be approximated well by models solely using the input question text. This enables preemptive filtering of questions that are not answered by the system due to their answer confidence scores being lower than the system threshold. Specifically, we learn Transformer-based question models by distilling Transformer-based answering models. Our experiments on three popular QA datasets and one industrial QA benchmark demonstrate the ability of our question models to approximate the Precision/Recall curves of the target QA system well. These question models, when used as filters, can effectively trade off lower computation cost of QA systems for lower Recall, e.g., reducing computation by ~60%, while only losing ~3-4% of Recall.

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。

↑