BriefGPT.xyz
Jun, 2021
了解问题更多：在问答中提高校准度
Knowing More About Questions Can Help: Improving Calibration in Question Answering
HTML
PDF
Shujian Zhang, Chengyue Gong, Eunsol Choi
TL;DR
研究了问题回答中的校准问题，通过利用输入示例信息和数据增强，提高了在阅读理解基准测试中校准准确性5-10％，并在开放检索设置中呈现了第一次的校准研究，比较了基于检索的跨度预测模型和答案生成模型的校准精度
Abstract
We study
calibration
in
question answering
, estimating whether model correctly predicts answer for each question. Unlike prior work which mainly rely on the model's confidence score, our calibrator incorporates i
→