BriefGPT.xyz
Feb, 2024
NoisyICL: 模型参数微噪音对上下文学习的校正
NoisyICL: A Little Noise in Model Parameters Calibrates In-context Learning
HTML
PDF
Yufeng Zhao, Yoshihiro Sakai, Naoya Inoue
TL;DR
通过在模型参数中引入随机噪声，我们提出了NoisyICL方法，以改善In-Context Learning的性能和校准，实验证明NoisyICL能够产生更准确、更公平、更可靠的预测结果。
Abstract
in-context learning
(ICL) is suffering from unsatisfactory
performance
and under-
calibration
due to high prior bias and unfaithful confide
→