BriefGPT.xyz
Oct, 2022
通过领域适应改善Prompt Tuning的样本效率
Improving the Sample Efficiency of Prompt Tuning with Domain Adaptation
HTML
PDF
Xu Guo, Boyang Li, Han Yu
TL;DR
本文提出了一种名为OPTIMA的算法,通过领域自适应来改进预处理语言模型的prompt tuning,结果表明OPTIMA可以显著提高prompt tuning的可迁移性和样本效率,并在少样本情况下超过全模型调整性能。
Abstract
prompt tuning
, or the conditioning of a frozen
pretrained language model
(PLM) with soft prompts learned from data, has demonstrated impressive performance on a wide range of
→