KG-FiD: Infusing Knowledge Graph in Fusion-in-Decoder for Open-Domain Question Answering
Donghan Yu1 , Chenguang Zhu2, Yuwei Fang2, Wenhao Yu , Shuohang Wang2, Yichong Xu2, Xiang Ren4, Yiming Yang1, Michael Zeng2 1Carnegie Mellon University 2Microsoft Cognitive Services Research Group 3University of Notre Dame 4University of Southern California Work done during internship at Microsoft.
Abstract
Current Open-Domain Question Answering (ODQA) models typically include a retrieving module and a reading module, where the retriever selects potentially relevant passages from open-source documents for a given question, and the reader produces an answer based on the retrieved passages. The recently proposed Fusion-in-Decoder (FiD) framework is a representative example, which is built on top of a dense passage retriever and a generative reader, achieving the state-of-the-art performance. In this paper we further improve the FiD approach by introducing a knowledge-enhanced version, namely KG-FiD. Our new model uses a knowledge graph to establish the structural relationship among the retrieved passages, and a graph neural network (GNN) to re-rank the passages and select only a top few for further processing. Our experiments on common ODQA benchmark datasets (Natural Questions and TriviaQA) demonstrate that KG-FiD can achieve comparable or better performance in answer prediction than FiD, with less than 40% of the computation cost.
中文速览
开放域问答（Open-Domain QA）系统通常先从海量文档中检索相关段落，再由阅读模块生成答案，而当前最先进的Fusion-in-Decoder（FiD）框架虽然效果出色，却面临两个痛点：各段落之间的关联信息被完全忽视，以及每个问题都要处理约100个段落导致计算量极大。为此，研究者提出KG-FiD，引入知识图谱（Knowledge Graph）来刻画段落间的结构关系，并用图神经网络（Graph Neural Network，GNN）对段落做两阶段重排序：第一阶段在检索后用DPR的段落向量初始化图节点、筛掉大量无关段落；第二阶段在编码器中间层提取表示、再次精排，只让少数高质量段落继续进入深层编码器和解码器生成答案。在Natural Questions和TriviaQA两个标准测试集上，KG-FiD以不足FiD 40%的计算量达到了相当甚至更好的答案预测效果，为在实际场景中部署高性能开放域问答系统提供了更高效的路径。
原文 arXiv:2110.04330;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2110.04330v2