BriefGPT.xyz
Nov, 2020
多选题问答系统的期望
What do we expect from Multiple-choice QA Systems?
HTML
PDF
Krunal Shah, Nitish Gupta, Dan Roth
TL;DR
本研究对最近在多项选择题回答（MCQA）数据集中取得高分的模型进行扰动实验，发现其表现不符合语言理解的人类期望，提出了一种新的训练方法，使模型更好地学习输入数据并使模型性能更好。
Abstract
The recent success of
machine learning
systems on various
qa datasets
could be interpreted as a significant improvement in models'
language under
→