Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/CLUEbenchmark/FewCLUE
FewCLUE 小样本学习测评基准,中文版
https://github.com/CLUEbenchmark/FewCLUE
benchmark bert chinese clue few-shot-learning fewclue gpt3 pet ptuning
Last synced: about 1 month ago
JSON representation
FewCLUE 小样本学习测评基准,中文版
- Host: GitHub
- URL: https://github.com/CLUEbenchmark/FewCLUE
- Owner: CLUEbenchmark
- Created: 2021-04-28T16:14:06.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2022-09-21T07:08:33.000Z (over 2 years ago)
- Last Synced: 2024-08-02T16:55:48.541Z (5 months ago)
- Topics: benchmark, bert, chinese, clue, few-shot-learning, fewclue, gpt3, pet, ptuning
- Language: Python
- Homepage: https://arxiv.org/abs/2107.07498
- Size: 59.8 MB
- Stars: 488
- Watchers: 13
- Forks: 72
- Open Issues: 8
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- StarryDivineSky - CLUEbenchmark/FewCLUE - shot Learning)正是解决这类在极少数据情况下的机器学习问题。结合预训练语言模型通用和强大的泛化能力基础上,探索小样本学习最佳模型和中文上的实践,是本课题的目标。FewCLUE:中文小样本学习测评基准,基于CLUE的积累和经验,并结合少样本学习的特点和近期的发展趋势,精心设计了该测评,希望可以促进中文领域上少样本学习领域更多的研究、应用和发展。模型有5种不同的方式做任务,分别是使用预训练模型直接做下游任务微调、PET、RoBERTa为基础的Ptuning方式、GPT类模型为基础的Ptuning方式、使用RoBERTa或GPT做零样本学习。 (A01_文本生成_文本对话 / 大语言对话模型及数据)