cloudyskyy / Guwen-UNILM
本仓库是基于bert4keras实现的古文-现代文翻译模型。具体使用了基于掩码自注意力机制的UNILM(Li al., 2019)预训练模型作为翻译系统的backbone。我们首先使用了普通的中文（现代文）BERT、Roberta权重作为UNILM的初始权重以训练UNILM模型（具体在文中分别为B-UNILM以及R-UNILM）。为了更好的使UNILM模型适应古文的特性，我们尝试使用了在古文预训练模型Guwen-BERT，作为UNILM的初始权重，并且获得了最优的效果。
☆46Updated 2 years ago
Alternatives and similar repositories for Guwen-UNILM:
Users that are interested in Guwen-UNILM are comparing it to the libraries listed below
- 古文语言理解测评基准 Classical Chinese Language Understanding Evaluation Benchmark: datasets, baselines, pre-trained models, corpus and leaderboard☆47Updated last year
- ☆33Updated 2 years ago
- A Benchmark for Classical Chinese Based on a Crowdsourcing System.☆55Updated 3 years ago
- SikuBERT：四库全书的预训练语言模型（四库BERT） Pre-training Model of Siku Quanshu☆121Updated last year
- 文言文命名实体识别，基于BILSTM+CRF完成文言文的命名实体实体，识别实体包括人物、地点、机构、时间等。☆9Updated 4 years ago
- GuwenModels: 古文自然语言处理模型合集, 收录互联网上的古文相关模型及资源. A collection of Classical Chinese natural language processing models, including Classical Ch…☆168Updated last year
- 一个面向繁体中文古籍分词的python工具包☆32Updated 3 years ago
- ☆25Updated last year
- a Corpus for Classical Chinese Language Event Extraction☆17Updated last year
- CCL2022汉语学习者文本纠错评测任务赛道二——CGED-8第一名解决方案☆52Updated last year
- 非官方的MDCSpell论文的实现☆18Updated 2 years ago
- This repository is intended for people who are interesting in learning/reading Classical Chinese (文言文) but be at a loss what to do to. As…☆12Updated 3 years ago
- 根据维基中文语料库预训练 GloVe 中文词向量；Pre-train GloVe word-embedding From Chinese Wiki corpus☆70Updated last year
- Dynamic Connected Networks for Chinese Spelling Check☆50Updated 10 months ago
- ChineseBert用于中文拼写纠错☆40Updated last year
- 古文现代文翻译平行语料库☆100Updated 3 years ago
- SIGHAN中文纠错数据集及转换后格式☆63Updated 5 years ago
- Yet Another Chinese Learner Corpus☆77Updated 3 years ago
- 大规模中文自然语言处理语料 Large Scale Chinese Corpus for NLP☆24Updated 6 years ago
- CCL 2022 汉语学习者文本纠错评测☆138Updated 2 years ago
- Ancient Chinese Corpus with Word Sense Annotation☆46Updated 8 months ago
- Finetune t5 and bart on Chinese Grammatical Error Correction data.☆17Updated 2 years ago
- 基于词汇信息融合的中文NER模型☆164Updated 2 years ago
- 基于GOOGLE T5中文生成式模型的摘要生成/指代消解，支持batch批量生成，多进程☆219Updated last year
- A Multi-modal Model Chinese Spell Checker Released on ACL2021.☆155Updated last year
- 基于pytorch的中文拼写纠错，使用的模型是Bert以及SoftMaskedBert☆31Updated 3 years ago
- ☆127Updated 2 years ago
- Extractive text summarization using seq2seq, PGN and coverage machanism.☆20Updated 2 years ago
- 复现论文《Simplify the Usage of Lexicon in Chinese NER》☆42Updated 3 years ago
- 历届中文句法错误诊断技术评测数据集☆38Updated 2 years ago