GRUE
收藏arXiv2023-03-01 更新2024-08-06 收录
下载链接:
http://arxiv.org/abs/2210.01241v3
下载链接
链接失效反馈官方服务:
资源简介:
GRUE是一个包含6个语言生成任务的基准,用于评估强化学习算法在自然语言处理任务中的表现。这些任务不是由目标字符串监督,而是由捕捉自动化人类偏好度量的奖励函数监督。
GRUE is a benchmark comprising six language generation tasks, designed to evaluate the performance of reinforcement learning algorithms in natural language processing tasks. These tasks are not supervised by target strings, but rather by reward functions that capture automated human preference metrics.
创建时间:
2022-10-04



