arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2004.00061cs.AIcs.CLcs.IR
收藏

基于统一的科学问题多跳推理的重构

Unification-based Reconstruction of Multi-hop Explanations for Science Questions

University of Manchester(曼彻斯特大学);Idiap Research Institute(Idiap研究所)

Marco Valentino, Mokanarangan Thayaparan, André Freitas

摘要:

提出了一种新的科学问答(QA)多跳解释重构框架。虽然现有的多跳推理方法建立解释考虑每个问题孤立,我们提出了一种方法来利用解释模式出现在一个语料库的科学解释。具体而言,该框架通过将词汇相关性与统一力的概念相结合来对一组原子事实进行排名,估计分析语料库中类似问题的解释。一个广泛的评价进行的Worldtree语料库,集成k-NN聚类和信息检索(IR)技术。我们提出以下结论:(1)所提出的方法实现了与Transformers竞争的结果,但速度快了几个数量级,这一特征使其可扩展到大型解释性语料库(2)基于统一的机制在减少语义漂移方面具有关键作用,有助于许多啤酒花解释的重建(6个或更多事实)和复杂推理事实的排名(+12.0平均平均精度)(3)至关重要的是,构建的解释可以支持下游QA模型,将BERT的准确性整体提高高达10%。

英文摘要:

This paper presents a novel framework for reconstructing multi-hop explanations in science Question Answering (QA). While existing approaches for multi-hop reasoning build explanations considering each question in isolation, we propose a method to leverage explanatory patterns emerging in a corpus of scientific explanations. Specifically, the framework ranks a set of atomic facts by integrating lexical relevance with the notion of unification power, estimated analysing explanations for similar questions in the corpus. An extensive evaluation is performed on the Worldtree corpus, integrating k-NN clustering and Information Retrieval (IR) techniques. We present the following conclusions: (1) The proposed method achieves results competitive with Transformers, yet being orders of magnitude faster, a feature that makes it scalable to large explanatory corpora (2) The unification-based mechanism has a key role in reducing semantic drift, contributing to the reconstruction of many hops explanations (6 or more facts) and the ranking of complex inference facts (+12.0 Mean Average Precision) (3) Crucially, the constructed explanations can support downstream QA models, improving the accuracy of BERT by up to 10% overall.

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。

↑