Texygen: A Benchmarking Platform for Text Generation Models

被引：240

作者：

Zhu, Yaoming ^{[1
]}

Lu, Sidi ^{[1
]}

Zheng, Lei ^{[1
]}

Guo, Jiaxian ^{[1
]}

Zhang, Weinan ^{[1
]}

Wang, Jun ^{[2
]}

Yu, Yong ^{[1
]}

机构：

[1] Shanghai Jiao Tong Univ, Shanghai, Peoples R China

[2] UCL, London, England

来源：

ACM/SIGIR PROCEEDINGS 2018 | 2018年

基金：

中国国家自然科学基金;

关键词：

Text Generation; Benchmarking; Evaluation Metrics;

D O I：

10.1145/3209978.3210080

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

We introduce Texygen, a benchmarking platform to support research on open-domain text generation models. Texygen has not only implemented a majority of text generation models, but also covered a set of metrics that evaluate the diversity, the quality and the consistency of the generated texts. The Texygen platform could help standardize the research on text generation and improve the reproductivity and reliability of future research work in text generation.

引用

页码：1097 / 1100

页数：4

共 16 条

[1] Abadi M., 2016, TENSORFLOW LARGESCAL
[2] [Anonymous], 2017, ARXIV170908624
[3] [Anonymous], 2015, ARXIV151105101
[4] Bahdanau D, 2016, Arxiv, DOI arXiv:1409.0473
[5] Che Tong, 2017, ARXIV170207983
[6] Generative Adversarial Networks
Goodfellow, Ian
Pouget-Abadie, Jean
Mirza, Mehdi
Xu, Bing
Warde-Farley, David
Ozair, Sherjil
Courville, Aaron
Bengio, Yoshua
[J]. COMMUNICATIONS OF THE ACM, 2020, 63 (11) : 139 - 144
[7] Jelinek Fred, 1977, JASA, V62, pS1
[8] Kusner M. J., 2016, ARXIV161104051
[9] Li J., 2017, P 2017 C EMP METH NA, P2157, DOI [10.18653/v1/D17-1230, DOI 10.18653/V1/D17-1230]
[10] Mikolov T, 2010, 11TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION 2010 (INTERSPEECH 2010), VOLS 1-2, P1045

← 1 2 →