Training a Ranking Function for Open-Domain Question Answering
Phu Mon Htut1、Samuel R. Bowman1,2,3、Kyunghyun Cho1,3 \AND1Center for Data Science New York University 60 Fifth Avenue New York, NY 10011、2Dept. of Linguistics New York University 10 Washington Place New York, NY 10003、3Dept. of Computer Science New York University 60 Fifth Avenue New York, NY 10011
Abstract
In recent years, there have been amazing advances in deep learning methods for machine reading. In machine reading, the machine reader has to extract the answer from the given ground truth paragraph. Recently, the state-of-the-art machine reading models achieve human level performance in SQuAD which is a reading comprehension-style question answering (QA) task. The success of machine reading has inspired researchers to combine information retrieval with machine reading to tackle open-domain QA. However, these systems perform poorly compared to reading comprehension-style QA because it is difficult to retrieve the pieces of paragraphs that contain the answer to the question. In this study, we propose two neural network rankers that assign scores to different passages based on their likelihood of containing the answer to a given question. Additionally, we analyze the relative importance of semantic similarity and word level relevance matching in open-domain QA.
中文速览
开放域问答（open-domain QA）系统无法像阅读理解那样直接拿到含有答案的段落，必须先从海量文本中检索出相关段落再提取答案，而现有检索步骤往往成为整个系统的瓶颈。针对这一问题，研究者设计了两种神经网络重排序模型：一种基于 InferSent 句向量，侧重捕捉问题与段落之间的语义相似度；另一种基于关系网络（Relation-Networks），侧重词级别的局部相关性匹配。在 QUASAR-T 数据集上的实验表明，关系网络排序器能检索到更多含答案的段落（召回率更高），但 InferSent 排序器最终带来了更好的问答准确率，将整体 QA 性能提升了 11.6%。这一结论揭示了语义相似度与词级相关匹配在开放域问答中各自扮演的不同角色，为后续检索增强型问答系统的设计提供了重要参考。
原文 arXiv:1804.04264;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1804.04264v1