生成式文本质量的自动评估方法综述(A Survey of Automatic Evaluation on the Quality of Generated Text)

Lan Tian (兰天), Ma Ziao (马梓奥), Zhou Yanghao (周杨浩), Xu Chen (徐晨), Mao Xianling (毛先领)


Abstract
“人工评估,作为生成式文本质量评价的金标准,成本太高;自动评估,核心思想在于要使其评估结果与人工评估高度相关,从而实现对生成式文本质量的自动化分析和评价。随着自然语言处理领域相关技术的迭代进步,使得生成式文本质量的自动评估技术,已然经历了多次技术范式的迭代。然而,学界至今依然缺乏对生成式文本质量自动评估技术的系统化总结。因此,本文将首先系统地对已有的生成式文本自动评估方法进行归纳总结,然后分析了生成式文本自动评估方法的主要发展趋势,最后为了使读者更加宏观地了解自动评估整体,对自动评估领域整体的未来研究方向进行了探讨和展望。”
Anthology ID:
2024.ccl-2.10
Volume:
Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 2: Frontier Forum)
Month:
July
Year:
2024
Address:
Taiyuan, China
Editor:
Xin Zhao
Venue:
CCL
SIG:
Publisher:
Chinese Information Processing Society of China
Note:
Pages:
169–196
Language:
Chinese
URL:
https://aclanthology.org/2024.ccl-2.10/
DOI:
Bibkey:
Cite (ACL):
Lan Tian, Ma Ziao, Zhou Yanghao, Xu Chen, and Mao Xianling. 2024. 生成式文本质量的自动评估方法综述(A Survey of Automatic Evaluation on the Quality of Generated Text). In Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 2: Frontier Forum), pages 169–196, Taiyuan, China. Chinese Information Processing Society of China.
Cite (Informal):
生成式文本质量的自动评估方法综述(A Survey of Automatic Evaluation on the Quality of Generated Text) (Tian et al., CCL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.ccl-2.10.pdf