Knowledge Fusion and Semantic Knowledge Ranking for Open Domain Question Answering
Pratyay Banerjee Chitta Baral Affiliation: Department of Computer Science, Arizona State University Email:
Abstract
Open Domain Question Answering requires systems to retrieve external knowledge and perform multi-hop reasoning by composing knowledge spread over multiple sentences. In the recently introduced open domain question answering challenge datasets, QASC and OpenBookQA, we need to perform retrieval of facts and compose facts to correctly answer questions. In our work, we learn a semantic knowledge ranking model to re-rank knowledge retrieved through Lucene based information retrieval systems. We further propose a “knowledge fusion model” which leverages knowledge in BERT-based language models with externally retrieved knowledge and improves the knowledge understanding of the BERT-based language models. On both OpenBookQA and QASC datasets, the knowledge fusion model with semantically re-ranked knowledge outperforms previous attempts.
中文速览
开放域问答的难点在于从海量知识中找出相关事实，并把分散在多句话里的信息组合起来回答科学选择题。研究者先用 Lucene/Elasticsearch 做多步检索，再用基于 BERT 的语义知识排序模型重新筛选事实，同时设计知识融合模块，把外部知识与语言模型自身的知识结合起来。实验显示，该方法在 OpenBookQA 和 QASC 上都超过了此前最好结果，准确率分别提升 2.2% 和 7.28%。这说明仅靠表面匹配难以完成多跳推理，而“先检索、再语义筛选、最后融合推理”的流程能更有效地利用外部知识。
原文 arXiv:2004.03101;中英对照 + 大白话阅读 https://aha.fim.ai/paper/2004.03101v2