BriefGPT.xyz
Nov, 2022
使用ReLM验证大型语言模型
Validating Large Language Models with ReLM
HTML
PDF
Michael Kuchnik, Virginia Smith, George Amvrosiadis
TL;DR
本文介绍了用标准正则表达式验证和查询LLMs的ReLM系统,其结果表明ReLM的最高系统效率是现有方法的15倍,数据效率是现有方法的2.5倍,并且具有竞争力和较高的统计学和提示调整覆盖率。
Abstract
Although
large language models
(LLMs) have been touted for their ability to generate natural-sounding text, there are growing concerns around possible negative effects of LLMs such as data
memorization
,
→