arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.07598cs.CL
收藏

面向多跳问答的多步以语义为中心的信息检索

Multi-step Entity-centric Information Retrieval for Multi-Hop Question Answering

University of Massachusetts Amherst(马萨诸塞大学阿默斯特分校);IBM Research(IBM研究院);Google Research(谷歌研究院)

Ameya Godbole, Dilip Kavarthapu, Rajarshi Das, Zhiyu Gong, Abhishek Singhal, Hamed Zamani, Mo Yu, Tian Gao, Xiaoxiao Guo, Manzil Zaheer, Andrew McCallum

摘要:

多跳问答(QA)需要一个信息检索(IR)系统,该系统可以找到回答问题所需的支持证据\emph{multiple},这使得检索过程非常具有挑战性。本文介绍了一种IR技术,它使用的信息实体存在于最初检索的证据,学习“\emph{hop}”到其他相关的证据。在设置中,超过\textbf{5 million}维基百科的段落,我们的方法导致检索性能显着提高。检索到的证据还提高了现有QA模型(未经任何培训)在\textbf{10.59} F1的\hotpot基准测试中的性能。

英文摘要:

Multi-hop question answering (QA) requires an information retrieval (IR) system that can find \emph{multiple} supporting evidence needed to answer the question, making the retrieval process very challenging. This paper introduces an IR technique that uses information of entities present in the initially retrieved evidence to learn to `\emph{hop}' to other relevant evidence. In a setting, with more than \textbf{5 million} Wikipedia paragraphs, our approach leads to significant boost in retrieval performance. The retrieved evidence also increased the performance of an existing QA model (without any training) on the \hotpot benchmark by \textbf{10.59} F1.

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。

↑