模型:
hfl/chinese-electra-small-discriminator
请在重新训练这些模型时使用ElectraForPreTraining作为鉴别器,使用ElectraForMaskedLM作为生成器。
谷歌和斯坦福大学发布了一个新的预训练模型,称为ELECTRA,与BERT及其变种相比,模型尺寸更小,性能相对竞争力强。为了进一步推动中国预训练模型的研究,哈工大与讯飞联合实验室(HFL)基于ELECTRA的官方代码发布了中文ELECTRA模型。ELECTRA-small只需BERT及其变种参数的1/10,就能在若干自然语言处理任务中达到或甚至超过相似的分数。
该项目基于ELECTRA的官方代码: https://github.com/google-research/electra
您可能还对以下内容感兴趣,
HFL的更多资源: https://github.com/ymcui/HFL-Anthology
如果您觉得我们的资源或论文对您有用,请在您的论文中包含以下引用。
@inproceedings{cui-etal-2020-revisiting, title = "Revisiting Pre-Trained Models for {C}hinese Natural Language Processing", author = "Cui, Yiming and Che, Wanxiang and Liu, Ting and Qin, Bing and Wang, Shijin and Hu, Guoping", booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings", month = nov, year = "2020", address = "Online", publisher = "Association for Computational Linguistics", url = "https://www.aclweb.org/anthology/2020.findings-emnlp.58", pages = "657--668", }