BriefGPT.xyz
Apr, 2021
QA-GNN: 用语言模型和知识图谱进行问答推理
QA-GNN: Reasoning with Language Models and Knowledge Graphs for Question Answering
HTML
PDF
Michihiro Yasunaga, Hongyu Ren, Antoine Bosselut, Percy Liang, Jure Leskovec
TL;DR
该研究提出了一种新模型QA-GNN，通过关联系统知识图和语言模型，实现对问题与答案之间的共同推理，并取得了在常识和生物医学领域QA测试中优于现有模型的表现。
Abstract
The problem of answering questions using knowledge from pre-trained
language models
(LMs) and
knowledge graphs
(KGs) presents two challenges: given a QA context (question and answer choice), methods need to (i) i
→