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 2304.01003cs.CLcs.LG
收藏

QUADRo:面向问答数据库检索的数据集与模型

QUADRo: Dataset and Models for QUestion-Answer Database Retrieval

University of Trento(特伦托大学);Amazon Alexa AI(亚马逊Alexa人工智能部门)

Stefano Campese, Ivano Lauriola, Alessandro Moschitti

AI总结:

本文提出QUADRo,一个用于开放领域问答数据库检索的大规模数据集与模型,通过神经信息检索和问答对重排序器,证明基于Transformer的模型优于仅依赖问题表示的模型,并与基于Web的方法竞争力相当。

摘要:

构建自动问答系统的一种有效范式是复用已回答的问题,例如用于FAQ或论坛应用。给定一个包含问答(q/a)对的数据库(DB),可以通过扫描DB中的相似问题来回答目标问题。在本文中,我们将该方法扩展到开放领域,使其与基于非结构化文档或图的其他标准方法具有竞争力。为此,我们(i)使用公开问题构建了一个包含630万个q/a对的大规模DB,(ii)设计了一个基于神经信息检索(IR)和q/a对重排序器的新系统,并(iii)构建了训练和测试数据,以对我们的模型进行对比实验。我们证明,对于神经搜索和重排序,使用(q,a)对的基于Transformer的模型优于仅基于问题表示的模型。此外,我们表明基于DB的方法与基于Web的方法(即建立在BING搜索引擎之上的QA系统)具有竞争力,这表明了寻找相关信息的挑战性。最后,我们公开了数据和模型以供未来研究使用。

英文摘要:

An effective paradigm for building Automated Question Answering systems is the re-use of previously answered questions, e.g., for FAQs or forum applications. Given a database (DB) of question/answer (q/a) pairs, it is possible to answer a target question by scanning the DB for similar questions. In this paper, we scale this approach to open domain, making it competitive with other standard methods, e.g., unstructured document or graph based. For this purpose, we (i) build a large scale DB of 6.3M q/a pairs, using public questions, (ii) design a new system based on neural IR and a q/a pair reranker, and (iii) construct training and test data to perform comparative experiments with our models. We demonstrate that Transformer-based models using (q,a) pairs outperform models only based on question representation, for both neural search and reranking. Additionally, we show that our DB-based approach is competitive with Web-based methods, i.e., a QA system built on top the BING search engine, demonstrating the challenge of finding relevant information. Finally, we make our data and models available for future research.

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。

↑